Abstract
We study the decay of the covariance of the \(\hbox {Airy}_1\) process, \({{\mathcal {A}}}_1\), a stationary stochastic process on \(\mathbb {R}\) that arises as a universal scaling limit in the Kardar–Parisi–Zhang (KPZ) universality class. We show that the decay is super-exponential and determine the leading order term in the exponent by showing that \({{\,\textrm{Cov}\,}}({{\mathcal {A}}}_1(0),\mathcal{A}_1(u))= e^{-(\frac{4}{3}+o(1))u^3}\) as \(u\rightarrow \infty \). The proof employs a combination of probabilistic techniques and integrable probability estimates. The upper bound uses the connection of \(\mathcal{A}_1\) to planar exponential last passage percolation and several new results on the geometry of point-to-line geodesics in the latter model which are of independent interest; while the lower bound is primarily analytic, using the Fredholm determinant expressions for the two point function of the \(\hbox {Airy}_1\) process together with the FKG inequality.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction and the Main Result
The one-dimensional Kardar–Parisi–Zhang (KPZ) universality class [40] of stochastic growth models has received a lot of attention in recent years, see e.g. the surveys and lecture notes [21, 25, 29, 33, 49, 50, 56, 59]. Two of the most studied models in this class are the exponential/geometric last passage percolation (LPP) and the totally asymmetric simple exclusion process (TASEP). In both cases, one can define a height function h(x, t), where x stands for space (one-dimensional in our case) and t for time.
At a large time t, under the \(2/3-1/3\) scaling, one expects to see a non-trivial limit process. To illustrate it, consider the scaling around the origin
with \(h_{\textrm{ma}}(\xi )=\lim _{t\rightarrow \infty } t^{-1} h(\xi t)\) being the (deterministic) macroscopic limit shape.
The limit process depends on the geometry of the initial condition. One natural initial condition is the stationary one and the limit process in this case, called \(\hbox {Airy}_{\textrm{stat}}\), has been determined in [6]. For non-random initial conditions, the two main cases are:
-
(a)
curved limit shape \(h_{\textrm{ma}}\): one expects the weak limit \(\lim _{t\rightarrow \infty } h^{\textrm{resc}}_t(u)=a_1 {{\mathcal {A}}}_2(a_2 u)\), with \({{\mathcal {A}}}_2\) being the \(\hbox {Airy}_2\) process [48] and \(a_1,a_2\) are model-dependent parameters (see [17, 38, 48] for LPP and TASEP setting and [27] for a non-determinantal case),
-
(b)
flat limit shape \(h_{\textrm{ma}}\): one expects the weak limit \(\lim _{t\rightarrow \infty } h^{\textrm{resc}}_t(u)=a_1' {{\mathcal {A}}}_1(a_2' u)\), with \({{\mathcal {A}}}_1\) being the \(\hbox {Airy}_1\) process [52], with again \(a_1',a_2'\) model-dependent parameters (see [18, 19, 52]).
As universal limit objects in the KPZ universality class, the \(\text {Airy}_{\text {stat}}\), as well as \({{\mathcal {A}}}_1\) and \({{\mathcal {A}}}_2\) (which also are stationary stochastic processes in \(\mathbb {R}\)) have attracted much attention. It is known that the one point marginal for \({{\mathcal {A}}}_2\) is the GUE Tracy–Widom distribution from random matrix theory [48], whereas the one point marginal for \({{\mathcal {A}}}_1\) is a scalar multiple of the GOE Tracy–Widom distribution [19, 52]. The next fundamental question is naturally to understand the two point functions for these processes. Although there are explicit formulae available for the multi-point distributions, extracting asymptotics from these complicated formulae is non-trivial. Widom in [57] (see also [2] for a conditional result, and [54] for higher order expansion) showed thatFootnote 1
The extended correlation kernel of the \(\hbox {Airy}_1\) and the \(\hbox {Airy}_2\) processes have algebraically many similarities (see the review [28]). This allows, following the method used by Widom [57] to get an expression for the joint distribution of \({{\mathcal {A}}}_1(0)\) and \({{\mathcal {A}}}_2(u)\) in terms of a scalar Fredholm determinant. For the \(\hbox {Airy}_2\) process case, due to identities in terms of the Hasting-McLeod solution of Painlevé II equation, one can get exact expressions for the terms in inverse powers of u (see [54] for expansions up to \(u^{-10}\)). This is not the case for the \(\hbox {Airy}_1\) process and thus the question of understanding the decay of correlations in the \(\hbox {Airy}_1\) process had remained open until now.
A numerical study [16] clearly showed that the decay of the covariance for the \(\hbox {Airy}_1\) process is very different from that of \(\hbox {Airy}_2\), in that it decays super-exponentially fast, i.e., \(-\ln {{\,\textrm{Cov}\,}}({{\mathcal {A}}}_1(0),{{\mathcal {A}}}_1(u)) \sim u^{\delta }\) for some \(\delta >1\). Unfortunately, the numerical data of [16] are coming from a Matlab program developed in [15] and uses the 10-digits machine precision. From the data it was not possible to conjecture the true value of \(\delta \).
The reason behind the difference in the decay of the covariances of \({{\mathcal {A}}}_2\) and \({{\mathcal {A}}}_1\) can be explained as follows. In the curved limit shape situation, the space-time regions which essentially determine the values of h(0, t) and \(h(u t^{2/3},t)\) have an intersection whose size decays polynomially in u. In contrast, for the flat limit shape, except on a set whose probability goes to zero super-exponentially fast in u, these regions are disjoint.
The goal of this paper is to prove that the decay of covariance for the \(\hbox {Airy}_1\) process is super-exponential with \(\delta =3\). More precisely, we prove upper and lower bounds of the covariance where exponents have a matching leading order term. The following theorem is the main result of this paper.
Theorem 1.1
There exist constants \(c,c'>0\) such that for \(u>1\)
Clearly, the threshold \(u>1\) above is arbitrary, and by changing the constants \(c,c'\) we can get the same bounds for any u bounded away from 0.
The upper and the lower bounds in Theorem 1.1 are proved separately with very different arguments.
In Sect. 2 we prove the upper bound; see Corollary 2.2. For this purpose we consider the point-to-line exponential last passage percolation (LPP) which is known to converge to the \(\hbox {Airy}_1\) process under an appropriate scaling limit. Corollary 2.2 is an immediate consequence of Theorem 2.1 which proves the corresponding decorrelation statement in the LPP setting. The strategy for the upper bound follows the intuition that the decorrelation comes from the fact that the point-to-line geodesics for two initial points far from each other use mostly disjoint sets of random variables. To make this precise, we prove results that are of independent interest. In Theorem 2.3 we get an upper bound on probability of the transversal fluctuations of a point-to-line geodesic, while in Theorem 2.7 we give an upper bound on the probability of intersection of two point-to-line geodesics. The proof is mainly based on probabilistic arguments, but uses one point moderate deviation estimates for the point-to-point and point-to-line exponential LPP with optimal exponents. Such results have previously been proved in [44], but an estimate with the correct leading order term in the upper tail exponent is required for our purposes. These are obtained in Lemma A.3 and Lemma A.4 by using asymptotic analysis.
In Sect. 3 we prove the lower bound; see Theorem 3.1. For the lower bound we start with Hoeffding’s covariance formula, which says that the covariance of two random variables is given by the double integral of the difference between their joint distribution and the product of the two marginals; see (3.2). The joint distribution of the \(\hbox {Airy}_1\) process is given in terms of a Fredholm determinant (see (3.10)) and the proof uses analytic arguments to obtain precise estimates for these Fredholm determinants. A crucial probabilistic step here, however, is the use of the FKG inequality applied in the LPP setting, which, upon taking an appropriate scaling limit yields that the aforementioned integrand is always non-negative; see Lemma 3.2. This allows one to lower bound the covariance by estimating the integrand only on a suitably chosen compact set, which nonetheless leads to a lower bound with the correct value of the leading order exponent.
We finish this section with a brief discussion of some related works. Studying the decay of correlations in exponential LPP has recently received considerable attention. Following the conjectures in the partly rigorous work [34], the decay of correlations in the time direction has been studied for the stationary and droplet initial conditions in [31], where precise first order asymptotics were obtained (see also [10, 32] for works on the half-space geometry). Similar, but less precise, estimates for the droplet and flat initial conditions were obtained in [11, 12]. All these works also rely on understanding the localization and geometry of geodesics in LPP, some of those results are also useful for us. The lower bounds in [11, 12] also use the FKG inequality in the LPP setting and provide bounds valid in the pre-limit. One might expect that similar arguments can lead to a bound similar, but quantitatively weaker, to Theorem 3.1 valid in the LPP setting.
2 Upper Bound
In this section we prove the upper bound of Theorem 1.1. Along the way, we prove results of independent interest. Theorem 2.3 gives an upper bound of the probability of a point-to-line geodesic deviating from its characteristic. Theorem 2.7 gives an upper bound on the probability that two point-to-line geodesics starting away from each other intersect. The leading terms are expected to be optimal.
2.1 Last passage percolation setting
We consider exponential last passage percolation (LPP) on \(\mathbb {Z}^2\). Let \(\omega _{i,j}\sim \exp (1)\), \(i,j\in \mathbb {Z}\), be independent exponentially distributed random variables with parameter 1. For points \(u,v\in \mathbb {Z}^2\) with \(u\prec v\), i.e., \(u_1\le v_1\) and \(u_2\le v_2\), we denote the passage time between the points u and v by
where the maximum is taken over all up-right paths from u to v in \(\mathbb {Z}^2\). Denote by \(\Gamma _{u,v}\) the geodesic from u to v, that is, the path \(\pi \) maximizing the above sum. Furthermore, let \(\mathcal {L}_n=\{(x,y)\in \mathbb {Z}^2\,|\, x+y=n\}\) and denote by
the point-to-line last passage time, where the maximum is taken over all up-right paths going from u to \(\mathcal {L}_n\).
Let us mention some known limiting results of exponential LPP. LetFootnote 2
and define the rescaled LPP
Then, by the result on TASEP with density 1/2 [19, 52] which can be transferred to LPP using slow decorrelation [26], we know that
where \({{\mathcal {A}}}_1\) is the \(\hbox {Airy}_1\) process, in the sense of finite-dimensional distributions. Similarly, (see [38] for the geometric case and [22] for a two-parameter extension)
with \({{\mathcal {A}}}_2\) is the \(\hbox {Airy}_2\) process [48], where in [39] the convergence is weak convergence on compact sets.
We also denote by \(\Gamma _N\) (resp. \(\Gamma ^*_N\)) the (almost surely unique) geodesic attaining \(L_{(0,0),(N,N)}\) (resp. \(L_{(0,0),\mathcal {L}_{2N}}\)). For a directed path \(\pi \), we denote by \(\pi (t)=(x-y)/2\) where (x, y) is the unique point (if it exits) where \(\pi \) intersect \(\mathcal {L}_t\). The parameter t will be thought of as time and \(\pi (t)\) will be the position of the path at time t. For instance, if \(\Gamma ^*_N\) ends at J(u), then \(\Gamma ^*_N(2N)=u(2N)^{2/3}\), see also Fig. 1. We shall also denote by \(L(\pi )\) the passage time of the path \(\pi \), i.e., the sum of the weights on \(\pi \).
The rescaled last passage times \(L_N^*(u)\) and \(L_N(u)\) have super-exponential upper and lower tails (see e.g. Appendices of [31] for a collection of such results and references), which implies that the limit of their covariance is the covariance of their limit, i.e.,
So if \(\textrm{Cov}\left( L_N^*(u),L_N^*(0)\right) \sim e^{-\beta u^3}\), then \(\textrm{Cov}\left( {{\mathcal {A}}}_1(u),{{\mathcal {A}}}_1(0)\right) \sim e^{-4 \beta u^3}\).
We first state the upper bound on \(\textrm{Cov}\left( L_N^*(u),L_N^*(0)\right) \) which is the main result in this section.
Theorem 2.1
For \(N^{1/14}\gg u>1\),
for some \(c>0\).
The following corollary, proving the upper bound in Theorem 1.1, is immediate from (2.7) and the above theorem.
Corollary 2.2
For \(u>1\), we have
for some \(c>0\).
Before proceeding further, let us explain the heuristic idea behind the proof of Theorem 2.1. Let T denote the straight line joining I(u/2) and J(u/2). Let \({\tilde{L}}_{N}(0)\) (resp. \({\tilde{L}}_{N}(u)\)) denote the rescaled last passage time from (0, 0) to \(\mathcal {L}_{2N}\) (resp. from I(u) to \(\mathcal {L}_{2N}\)) in the LPP restricted to use the randomness only to the left (resp. to the right) of T. Since \({\tilde{L}}_{N}(0)\) and \({\tilde{L}}_{N}(u)\) depend on disjoint sets of vertex weights and hence are independent, one expects that the leading order behaviour of the covariance is given by the probability that \(L_N^*(0)\ne {\tilde{L}}_{N}(0)\) and \(L_N^*(u)\ne {\tilde{L}}_{N}(u)\) (parts of the sample space where only one of these two events hold can also contribute, but our arguments will show that these contributions are not of a higher order, see Fig. 1). Now,
and the probability of the last event is \(\lesssim e^{-\frac{1}{6}u^3}\) by Theorem 2.3 below. The proof is completed by showing that the probability of the intersection of the two events has an upper bound which is of the same order (at the level of exponents) as their product. This final step is obtained by considering the two cases, one where the point-to-line geodesics do not intersect, and the second where they do. The first part is bounded using the BK inequality where the probability of the geodesics intersecting is upper bounded separately in Theorem 2.7.
2.2 Localization estimates of geodesics
As explained above, a key step in the proof is to get precise estimates for the probability that the geodesic behaves atypically, i.e., it exits certain given regions. To this end, the main result of this subsection provides the following localization estimate for \(\Gamma ^*_N\) that is of independent interest.
Theorem 2.3
For \(N^{1/14}\gg u>1\) we have
for some constant \(c>0\).
Although we do not prove a matching lower bound, the constant \(\frac{4}{3}\) is expected to be optimal; see the discussion following Lemma 2.5. Transversal fluctuation estimates for geodesics in LPP are of substantial interest and have found many applications. This is a first optimal upper bound in this direction for point-to-line geodesic. For point-to-point geodesics, similar estimates (albeit with unspecified constants in front of the cubic exponent) are proved for Poissonian LPP [14] and exponential LPP [12], see also [35] for a lower bound. In fact, we shall need to use the following estimate from [12, 23].
Lemma 2.4
(Proposition C.9 of [12]). For \(N^{1/3}\gg u>1\) we have
for some constant \(c>0\).
For the proof of Theorem 2.3 as well as the results in the subsequent subsections of this section, we shall use as input some lower and upper tail estimates of various LPP, which are collected and, if needed, proved in Appendix A.2.
The first step is to prove the special case \(t=2N\) of Theorem 2.3; we get an estimate of the probability that \(\Gamma ^*_N\) ends in \({{\mathcal {D}}}_u=\cup _{v\ge u} J(v)\), that is, \(\Gamma ^*_N(2N)\ge u(2N)^{2/3}\).
Lemma 2.5
For all \(N^{1/9}\gg u>1\), we have
for some \(c>0\).
Again, we do not prove a matching lower bound, but the constant \(\frac{4}{3}\) should be optimal. Indeed, notice that by (2.6), one expects that \((2N)^{-2/3}\Gamma ^{*}_{N}(2N)\) weakly converges to the almost surely unique maximizer \({{\mathcal {M}}}\) of \({{\mathcal {A}}}_{2}(u)-u^2\). The distribution of \({{\mathcal {M}}}\) has been studied in [7, 53] whence it is known that \({{\,\mathrm{\mathbbm {P}}\,}}(|{{\mathcal {M}}}|\ge u)\sim e^{-\frac{4}{3}u^3}\).
Proof of Lemma 2.5
Let \(C_0\) be such that (using Lemma A.1)
We have
Using our definition of \(C_0\) and Lemma A.3, we get for \(N^{1/3}\gg u\ge C_0+1\)
By adjusting the constant c if necessary, we get the same conclusion for \(u\in [1,C_0+1]\).
\(\square \)
Our next result is a similar localization estimate for the point-to-line geodesic \(\Gamma ^*_N\) at an intermediate time t.
Lemma 2.6
Let \(t=2\tau N\). Fix any \(\theta \ge 1\) and take \(\min \{\tau ,1-\tau \}\ge u^{-\theta }\). Assume that \(N^{1/(9+3\theta )}\gg u>\max \{1,\tfrac{4}{3}\sqrt{\tau }C_0\}\) with \(C_0\) as in (2.13). Then there exists a constant \(c>0\) such that
In particular, by choosing another constant \(c'>0\),
for all \(N^{1/(9+3\theta )}\gg u>1\).
Notice that (2.16) provides a stronger bound compared to (2.17) for small \(\tau \) which is expected as the transversal fluctuation should grow with \(\tau \). Indeed, for \(\tau \ll 1\), one expects even stronger bounds, see Theorem 3 of [13]. For the proof of Theorem 2.3, (2.17) would suffice but we record the stronger estimate (2.16) as it would be used in the next subsection.
Proof of Lemma 2.6
We start with a simple rough bound. Notice that since the geodesics are almost surely unique, by planarity they cannot cross each other multiple times. Consequently, if \(\Gamma ^*_N(2N)\le A(2N)^{3}\) for some \(A>0\), the geodesic \(\Gamma ^{*}_{N}\) lies to the left of the point-to-point geodesic from I(A) to J(A). Therefore, the maximal transversal fluctuation of the point-to line geodesic can be upper bounded by the sum of the fluctuation at the endpoint plus the maximal transversal fluctuation of a point-to-point geodesic. Similar arguments will be used multiple times in the sequel and will be referred to as the ordering of geodesics. It follows that
Applying the bounds of Lemma 2.5 and Lemma 2.4 with the constant \(C_1\) large enough, we get that
Next we need to bound the probability that \(\Gamma ^*_N(t)\) is in \([u(2N)^{2/3},C_1u\tau ^{-1/2}(2N)^{2/3}]\). Define \(K(v)=(t/2,t/2)+v(2N)^{2/3}(1,-1)\). Then, for any \(S\in \mathbb {R}\),
where \({{\tilde{L}}}_{K(v),\mathcal {L}_{2N}}=L_{K(v),\mathcal {L}_{2N}}-\omega _{K(v)}\) is the LPP without the first pointFootnote 3. We set
By Corollary A.2, setting \(a=C_0 \tau ^{-1/2}u^{-1}\ll N^{2/3}\) with \(C_0\) as in (2.13), we see that
It remains to bound the last term in (2.20). We have
where \(\delta >0\) will be chosen later. Note that the two supremums above are independent. Denoting \(X_k=\sup _{v\ge u+k\delta } L_{(0,0),K(v)}\) and \(Y=\sup _{-\delta /2\le v\le \delta /2}\tilde{L}_{K(v),\mathcal {L}_{2N}}\), note also that \(\sup _{u+k\delta \le v\le u+(k+1)\delta }{{\tilde{L}}}_{K(v),\mathcal {L}_{2N}}\) has the same law as Y for each k. Hence,
since \(X_k\le X_0\) for any \(k\ge 0\).
As \(X_0\) and Y are independent, it is expected that the leading term should behave as
with \(S^*\) chosen such that (2.25) is maximal. Since we want to minimize over a finite number of points (not going to infinity as N does), we instead look for the maximum of
for some small positive discretization step \(\eta \); see Fig. 2. The natural scale of the fluctuation of \(X_0\) is \(\tau ^{1/3} 2^{4/3}N^{1/3}\) and the one for Y is \((1-\tau )^{1/3}2^{4/3}N^{1/3}\), so we choose \(\eta \) at most \(\tfrac{1}{4}\min \{(1-\tau )^{1/3},\tau ^{1/3}\}\).
We want to discretize the interval \([4\tau N-\frac{u^2}{\tau }2^{4/3}N^{1/3},4\tau N-a u^2 2^{4/3}N^{1/3}]\) into pieces of size \(\eta 2^{4/3} N^{1/3}\). The interval shall be non-empty, which can be ensured if u is not too small. For that purpose let us assume that \(u>\max \{1,\frac{4}{3} \sqrt{\tau }C_0\}\). Define the number of discretized points and then \(\eta \) by
Our assumption on u ensures that \(\frac{1}{\tau }-a\ge \frac{1}{4\tau }\), or equivalently \(\tau a\le \frac{3}{4}\), as well as \(\eta /u^2\le \frac{1}{4} 2^{-1/3}\). Therefore \(\tau a+\eta /u^2<1\).
Let \(S_1=4\tau N-\frac{u^2}{\tau }2^{4/3}N^{1/3}\) and \(S_2=S-S_1=4(1-\tau )N+\frac{1-\tau a}{\tau }u^2 2^{4/3}N^{1/3}\), \(S_3=4\tau N-a u^22^{4/3}N^{1/3}\). Let \(S^*\) be the maximizer of (2.26). Then
Instead of finding \(S^*\) that maximizes (2.26), we look to find an upper bound of (2.26) by maximizing the product of the upper bounds of the individual terms. For this, we take \(\delta =(1-\tau )^{2/3}\). Then by rescaling Lemma A.3, for \(s_2\ll (\tau N)^{2/9}\) and \(u\ll (\tau N)^{1/9}\)
and by rescaling Proposition A.5, for \(s_2\ll (1-\tau )^{2/3}N^{2/3}\), we get
Thus we need to maximize the product of the terms in (2.29) and (2.30). All the terms in the sum in (2.28) corresponds to \(s_1\) and \(s_2\) being non-negative and also of \(\mathcal {O}(\frac{u^2}{\tau \min \{\tau ^{1/3},(1-\tau )^{1/3}\}})\). Therefore we can ignore the polynomial pre-factor in (2.30) and look to find \(s_1,s_2\ge 0\) such that
Denote \({{\tilde{a}}}=a+\eta /u^2\), which satisfied \(\tau {{\tilde{a}}}<1\) by the above assumptions. Plugging in the value of \(s_2\) as function of \(s_1\) into \(s_1^{3/2}+s_2^{3/2}\) and computing its minimum through the derivatives we get
With this choice of \(s_1\) and \(s_2\), we write with a minor abuse of notation
and
for some constant \(C>0\), where we have used the a priori bound on \(s_2\) together with the assumption on \(\tau \) to get the polynomial pre-factor. For \(\tau {{\tilde{a}}} <1\), \(\left( \tau ^{-1}-\tilde{a}\right) ^{3/2}\ge \tau ^{-3/2}(1-\tfrac{3}{2} \tau a-\frac{3}{8} u^{-2})\) so that we get
This is the \(\tau \)-dependent bound which is useful for small enough \(\tau \), but the exponent is minimal when \(\tau \rightarrow 1\).
Finally, notice that the bound on \({{\,\mathrm{\mathbbm {P}}\,}}(X_0>S_3)\) (resp. \({{\,\mathrm{\mathbbm {P}}\,}}(Y>S_2)\)) corresponds to the bound with the value \(s_2=0\) (resp. \(s_1=0\)), thus they are also smaller than (2.34). Thus we have shown that (2.28)\(\le (M+2)\frac{C_1 u}{\sqrt{\tau }\delta } \times \) (2.35). The prefactor is only a polynomial in u (using again the assumptions on \(\tau \)) and can be absorbed in the \(u^2\) term in the exponent by adjusting the constant. Thus we have proved (2.16), and (2.17) follows by observing that \(\tau \rightarrow 1\) is the worst case. The condition \(u\ll N^{1/(9+3\theta )}\) ensures that all the conditions on \(s_1\), \(s_2\), u mentioned above are satisfied. \(\square \)
We can now prove Theorem 2.3.
Proof of Theorem 2.3
We shall prove the result for u sufficiently large, the result for all \(u>1\) shall follow by adjusting the constant c. Let us set \(\varepsilon =\delta u^{-3/2}\) for some \(\delta >0\) to be chosen later (\(\delta \) will be small but fixed and in particular will not depend on u or N). Without loss of generality let us also assume that \(\varepsilon N\) and \(1/\varepsilon \) are both integers. Let us define the sequence of points
Let \(A_{j}\) denote the event that \(\Gamma ^*_{N}(2j \varepsilon N)\ge (u-1)(2N)^{2/3}\) for \(j=1,2, \ldots , \varepsilon ^{-1}\), and let \(B_{j}\) denote the event that \(\sup _{t}\Gamma _{v_{j-1},v_{j}} \ge u (2N)^{2/3}\) for \(j=1,2,\ldots , \varepsilon ^{-1}\) where \(\Gamma _{v_{j-1},v_{j}}\) denotes the geodesic from \(v_{j-1}\) to \(v_{j}\). By ordering of geodesics, it follows that on
one has \(\sup _{0\le t\le 2N} \Gamma ^*_{N}(t) \le u(2N)^{2/3}\); see Fig. 3.
Hence,
It follows from Lemma 2.6 that for each j,
for some new constant \(c'>0\). Notice now that
We now use \(u\ll N^{2/3}\) (which implies \(u\ll (\varepsilon N)^{1/3}\)) and choose \(\delta \) sufficiently small such that by Lemma 2.4 we have
With this choice it follows that
Since \(\varepsilon ^{-1}=\mathcal {O}(u^{3/2})\), the result follows by adjusting the value of c. \(\square \)
2.3 Coalescence probability
Consider the point-to-line geodesics \(\Gamma ^*_{N}\) and \({\tilde{\Gamma }}^{*}_{N}\) to \(\mathcal {L}_{2N}\) from (0, 0) and I(u) respectively. Owing to the almost sure uniqueness of geodesics, if \(\Gamma ^*_{N}\) and \({\tilde{\Gamma }}^{*}_{N}\) meet, they coalesce almost surely. Coalescence of geodesics is an important phenomenon in random growth models including first and last passage percolation and has attracted a lot of attention. For exponential LPP, for point-to-point geodesics started at distinct points and ending at a common far away point (or semi-infinite geodesics going in the same direction) tail estimates for distance to coalescence has been obtained; see [13, 46, 58] for more on this. For point-to-line geodesics started at initial points that are far (on-scale), one expects the probability of coalescence to be small. Our next result proves an upper bound to this effect and is of independent interest.
Theorem 2.7
In the above set-up, for \(N^{1/14}\gg u>1\)
for some constant \(c>0\).
The rest of this section deals with the proof of Theorem 2.7. We divide it into several smaller results. As always, we shall assume without loss of generality that u is sufficiently large, extending the results to all \(u>1\) is achieved by adjusting constants.
First of all, due to Theorem 2.3, the probability that the two geodesics in the statement of Theorem 2.7 meet outside the rectangle \(\mathcal {R}(u)\) with corners (0, 0), I(u), J(u), (N, N) is smaller than the estimate we want to prove. Thus we can restrict to bounding the probability that the two geodesics intersect in \(\mathcal {R}(u)\); a stronger result is proved in Lemma 2.8 below. As the number of points in \(\mathcal {R}(u)\) where the geodesics \({\tilde{\Gamma }}^{*}_{N}\) and \(\Gamma ^*_N\) could meet is \(\mathcal {O}(N^{5/3})\), we need to discretize space. We therefore divide \(\mathcal {R}(u)\) into a grid of size \(\varepsilon N\times (2 \varepsilon N)^{2/3}\), where \(\varepsilon \) will be taken small enough (but not too small, namely of order \(u^{-2}\)); see Fig. 4.
For \(\tau \) an integer multiple of \(\varepsilon \), let \(A(\tau ,v)\) be the event that the first intersection of \(\Gamma ^*_N\) and \({\tilde{\Gamma }}^{*}_N\) occurs at time \(t \in (2\tau N,2\tau N+2\varepsilon N]\), and they then cross the anti-diagonal grid segment (of length \((2 \varepsilon N)^{2/3}\)) at time \(2\tau N+2\varepsilon N\) with mid-point given by
Notice that, the number of choices of \(\tau \) and v is \(\mathcal {O}(\varepsilon ^{-5/3})\), which by our choice of \(\varepsilon \) is at most a polynomial of u. Thus we need to prove the \({{\,\mathrm{\mathbbm {P}}\,}}(A(\tau ,v))\) is at most \(e^{-\frac{1}{3} u^3+c u^2}\) for any \(\tau ,v\). The proof of Theorem 2.7 is completed by taking a union bound.
Our first rough estimate deals with values of v which are close to 0 or u and also small values of \(\tau \). The basic idea is that in these cases the probability bounds coming from considering the transversal fluctuation of a single geodesic is sufficient.
Lemma 2.8
Let \(N^{1/14}\gg u >1\). For any v satisfying \(\min \{v,u-v\}\le u (1-2^{-2/3})-\varepsilon ^{2/3}\) and for any \(\tau \le 2^{-2/3}-\varepsilon \),
for some constant \(c>0\).
Proof
For \(u-v\le u(1-2^{-2/3})-\varepsilon ^{2/3}\), we have
where the last inequality follows from Theorem 2.3. The same argument gives the desired result for \(v\le (1-2^{-2/3})u\) by considering the transversal fluctuation of the geodesic \({\tilde{\Gamma }}_{N}^*\).
Next, notice that \(A(\tau ,v)\) implies that either \(\Gamma ^*_N(2\tau N+2\varepsilon N)\ge \frac{1}{2} u (2N)^{2/3}\) or \({\tilde{\Gamma }}^{*}_{N}(2\tau N+2\varepsilon N)\le \frac{1}{2} u (2N)^{2/3}\), since after meeting they follow the same path. For \(\tau +\varepsilon \le 2^{-2/3}\), by Lemma 2.6 (use the first inequality with \(\tau \mapsto \tau +\varepsilon \le 2^{-2/3}\)) each of these events have probability bounded by \(e^{-\frac{1}{3} u^3+c u^2}\) for some constant \(c>0\), completing the proof. \(\square \)
We now proceed towards dealing with the remaining case. Define the segment
Let \(C_2\) be large enough such that
For a path \(\gamma \), recall that \(L(\gamma )\) denotes the passage time of that path. Define the event
where in \({{\tilde{L}}}\) we remove the first point. Then we have the following estimate.
Lemma 2.9
Assume \((1-2^{-2/3})u-\varepsilon ^{2/3}<v<2^{-2/3}u+\varepsilon ^{2/3}\) and \(2^{-2/3}-\varepsilon \le \tau \le 1-\varepsilon \) (notice that if the geodesics coalesce then \(A(\tau ,v)\) must hold for some \(\tau \le 1-\varepsilon \), the case \(\tau =1\) need not be considered). For \(N^{1/11}\gg u>1\), there exists a \(\delta >0\) small enough (not depending on u and N) such that with \(\varepsilon =\delta u^{-2}\),
for some constant \(C>0\).
Proof
We prove it for \(u>4(1+C_2)\) with \(C_2\) as in (2.48). Then by adjusting the constant C it is true also for \(u>1\). Denote by \(\Gamma _1\) and \(\Gamma _2\) the geodesics from (0, 0) to \(P(\tau ,v-\frac{1}{2}\varepsilon ^{2/3})\) and from I(u) to \(P(\tau ,v+\frac{1}{2}\varepsilon ^{2/3})\) respectively. These two points are the end point of the grid interval whose midpoint is \(P(\tau ,v)\); see also Fig. 5. Define the events
Let us show that
Observe that
and the last event is included in \(B(\tau ,v)\). Indeed, on \(A(\tau ,v)\cap B_1^c\cap B_2^c\), the geodesics \(\Gamma ^*_N\) and \({\tilde{\Gamma }}^{*}_N\) must cross \({{\mathcal {S}}}_v\). Now, let \(\gamma _1\) and \(\gamma _2\) be the portions of \(\Gamma ^*_N\) and \({\tilde{\Gamma }}^{*}_N\) respectively before time \(\tau N\). On \(A(\tau ,v)\), by definition, \(\gamma _1\) and \(\gamma _2\) must be disjoint. On \(B_3^c\cap A(\tau ,v)\) it holds that \(L(\gamma _1)+L_{{{\mathcal {S}}}_v,\mathcal {L}_{2N}}\ge L_{(0,0),\mathcal {L}_{2N}}\ge 4N-C_2 u 2^{4/3}N^{1/3}\), and similar inequality holds on \(B_4^c \cap A(\tau ,v)\) replacing \(\gamma _1\) by \(\gamma _2\). Thus the event \(B(\tau ,v)\) is satisfied.
To complete the proof, we apply union bound to (2.52) and bound the probabilities \({{\,\mathrm{\mathbbm {P}}\,}}(B_i)\). By the choice of \(C_2\), \({{\,\mathrm{\mathbbm {P}}\,}}(B_3)\) and \({{\,\mathrm{\mathbbm {P}}\,}}(B_4)\) are both bounded by \(e^{-\frac{1}{3} u^3}\). Lemma 2.10 below shows that \({{\,\mathrm{\mathbbm {P}}\,}}(B_1)\le C e^{-\frac{1}{3} u^3}\) for some \(C>0\) and by symmetry \({{\,\mathrm{\mathbbm {P}}\,}}(B_2)\le C e^{-\frac{1}{3} u^3}\) as well. This completes the proof. \(\square \)
Lemma 2.10
Assume \((1-2^{-2/3})u-\varepsilon ^{2/3}<v<2^{-2/3}u+\varepsilon ^{2/3}\) and \(2^{-2/3}-\varepsilon \le \tau \le 1\). For \(N^{1/14}\gg u >1\), there exists a \(\delta >0\) small enough (not depending on u and N) such that with \(\varepsilon =\delta u^{-2}\),
for some \(C>0\).
Proof
\(\Gamma _1\) is the geodesic from (0, 0) to \(Q_1=P(\tau ,v-\frac{1}{2}\varepsilon ^{2/3})\). We want to bound the probability that \(\Gamma _1(2\tau N)\le (v-1)(2N)^{2/3}\).
For \(x_1,x_2,x_3>0\), define the events
By a first order approximation, we have \(L_{(0,0),Q_1}\simeq 4(\tau +\varepsilon )N-\frac{v^2}{\tau +\varepsilon }2^{4/3}N^{1/3}+\mathcal {O}(v\varepsilon ^{2/3}N^{1/3})\). So, by Lemma A.1, we have \({{\,\mathrm{\mathbbm {P}}\,}}(E_1)\le C e^{-c x_1^3}\) for \(x_1\) of at least the order of u, and such that \(x_1\ll N^{2/3}\). Next, by Lemma A.4, we have \({{\,\mathrm{\mathbbm {P}}\,}}(E_2)\le C e^{-\frac{4}{3} x_2^{3/2}\tau ^{-1/2}}\) for \(x_2\ll N^{2/9}\). Finally, using Lemma A.3 (with the variables (N, u) in Lemma A.3 replaced by \((\varepsilon N,\varepsilon ^{-2/3}(1-\tfrac{1}{2} \varepsilon ^{2/3}))\), we get \({{\,\mathrm{\mathbbm {P}}\,}}(E_3)\le C e^{-\frac{4}{3} x_3^{3/2}\varepsilon ^{-1/2}}\) provided
Under the condition
we have
We assume already that \(\varepsilon \) is small enough so that \(\tau \ge 1/2\). First take \(x_1=u/(3 c)\) so that \({{\,\mathrm{\mathbbm {P}}\,}}(E_1)\le C e^{-\frac{1}{3} u^3}\). Then take \(x_2=u^2\) which ensures \({{\,\mathrm{\mathbbm {P}}\,}}(E_2)\le C e^{-\frac{1}{3} u^3}\) as well. Finally we take \(x_3=u^2 \varepsilon ^{1/3}\) that gives \({{\,\mathrm{\mathbbm {P}}\,}}(E_3)\le C e^{-\frac{1}{3} u^3}\). To satisfy the condition (2.57), it is enough to take \(\varepsilon =\delta u^{-2}\) with \(\delta \) small enough (independent of u). Finally, note that (2.56) implies that \(u\ll N^{\tfrac{1}{14}}\). \(\square \)
To complete the proof of Theorem 2.7 we need to obtain a bound on the event \(B(\tau ,v)\).
Lemma 2.11
Assume \(2^{-2/3}u-\varepsilon ^{2/3}<v<(1-2^{-2/3})u+\varepsilon ^{2/3}\) and \(2^{-2/3}-\varepsilon \le \tau \le 1-\varepsilon \). For \(N^{1/9}\gg u>4(1+C_2)\), there exists a constant \(\delta >0\) small enough such that with \(\varepsilon =\delta u^{-2}\),
for some constant \(c>0\) independent of \(\tau ,v\). The constant \(C_2\) is as in (2.48).
Proof
For any \(s_1,s_2\in \mathbb {R}_+\cup \{-\infty \}\) we define \(D(s_1,s_2)\) to be the event that there exist disjoint paths \(\gamma _1\) and \(\gamma _2\) as in the definition of \(B(\tau ,v)\), such that
For \(s_3\in \mathbb {R}_+\cup \{-\infty \}\), we define the event
Recall the constant \(C_2\) from (2.48). Like in the proof of Lemma 2.6 we do a discretization with a fixed width \(0<\eta <1\) and thus we will not write all the details. The minor difference is that now we a couple of constraints:
In the discretization of Lemma 2.6, see (2.28), we separated explicitly two terms, which corresponds taking \(S_2=-\infty \) and \(S_3=-\infty \). Here we do the same, but instead of writing those terms separately, we consider subsets allowing positive numbers and \(-\infty \). More precisely, define the set
Then
The number of elements is, for any v with \(\min \{v,u-v\}\le u (1-2^{2/3})+\varepsilon ^{2/3}\) of order \(u^2/\tau \). Since \(\tau \ge 1/2\) (for \(\varepsilon \le 2^{-2/3}-1/2\)), the sum contains \(\mathcal {O}(u^2)\) many terms. Therefore, using the independence of \(C(s_3)\) and \(D(s_1,s_2)\),
As \(\gamma _1\) and \(\gamma _2\) “occur disjointly”, by the BK (Berg-Kesten) inequality (see e.g. Theorem 7 of [3] for a statement applicable in the above scenario) we get
Set \({{\mathcal {D}}}_{\tau ,v}=\{(\tau N+k,\tau N-k)| k\ge (v-1)(2N)^{2/3}\}\). Then \(L_{(0,0),{{\mathcal {S}}}_v}\le L_{(0,0),{{\mathcal {D}}}_{\tau ,v}}\) and Lemma A.3 (after rescaling \(N\rightarrow \tau N\)) leads to
and similarly for the bound on the second term in (2.66), so that we have
for \(2<u\ll N^{1/9}\), \(0<s_1,s_2\ll N^{2/9}\).
To estimate \({{\,\mathrm{\mathbbm {P}}\,}}(C(s_3))\) we divide the segment \({{\mathcal {S}}}_v\) into pieces of length \((1-\tau )^{2/3}(2N)^{2/3}\) to which we can apply a rescaled version of Proposition A.5. We have \(2/(1-\tau )^{2/3}\le 2/\varepsilon ^{2/3}\) such pieces (we used \(1-\tau \ge \varepsilon \)). Using union bound we then get, for \(s_3 (1-\tau )^{-1/3}\ll N^{2/3}\),
Let \(\varepsilon =\delta u^{-2}\) for \(\delta >0\) small enough as in the proof of Lemma 2.10. We have
Therefore we concentrate now on finding the maximum of
for \(s_1,s_2,s_3\in \Theta \). Define \({{\tilde{s}}}_3=s_3+\eta +C_2 u\). Then for given \(s_3\) on \(\Theta \), we have
So we need to maximize
In principle, to get the bound on \(B(\tau ,v)\), we would need to find \(s_3\) maximizing \(M(v,\tau ,s_3)\). In the statement we want a bound uniform in \(\tau ,v\). This means that we need to maximize the result over \(\tau ,v\) as well. In short, we maximize M for \(s_3,v,\tau \) and thus we do it in another order. First notice that for given \(\tau ,s_3\), \(M(v,\tau ,s_3)\) is maximized at \(v=u/2\), for which
Computing the derivative with respect to \(s_3\) we get that, for a given \(\tau \), the maximum is at \(s_3^*=\frac{[(u-2)^2-4\tau (\eta +C_2 u)](1-\tau )}{\tau (4-3\tau )}>0\) under the assumption \(u\ge 4(1+C_2)\) and \(\eta <1\). So we get
for some constant \(c>0\). Inserting (2.75) into (2.70) and choosing an appropriate new constant c leads to the claimed result. \(\square \)
2.4 Proof of Theorem 2.1
As before, we shall prove the bound first for sufficiently large u, and adjust c later to deduce the same for all \(u>1\).
Recall that \(L_N^*(u)\) is the rescaled LPP from I(u) to \(\mathcal {L}_{2N}\), see (2.4). Let us use the notations \(X=L_N^*(0)\) and \(Y=L_N^*(u)\). For \(j,j'\in \{1,2,\ldots , u-1\}\), let \(S^{j}_N\) denote the weight of the maximum weight path \(\pi ^j\) from (0, 0) to \(\mathcal {L}_{2n}\) such that \(\pi (t) < j(2N)^{2/3}\) for all t, and similarly, let \(S^{u,j'}_N\) denote the weight of the maximum weight path \({{\tilde{\pi }}}^{j'}\) from I(u) to \(\mathcal {L}_{2n}\) such that \({{\tilde{\pi }}}^{j'}(t) > (u-j')(2N)^{2/3}\) for all t. Let us also set
For notational convenience, we shall also write \(X_0=Y_0=0\) and \(X_{u}=X\), \(Y_{u}=Y\). Now, writing
and using the bilinearity of covariance it is enough to prove that for some \(c>0\) and for all \(j, j'\in \{0,1,\ldots , u-1\}\)
Notice first that \(X_{j+1}-X_{j}\) and \(Y_{j'+1}-Y_{j'}\) depend on disjoint sets of vertex weights and hence are independent unless \(j+j'\ge u-1\). Hence we only need to consider \((j,j')\) such that \(j+j'\ge u-1\). For such a pair, noticing \(X\ge X_{j+1}\ge X_{j}\) and \(Y\ge Y_{j+1}\ge Y_{j}\) it follows that
For convenience of notation, let \(\Gamma _1\) and \(\Gamma _2\) locally denote the geodesics from (0, 0) and I(u) respectively to \(\mathcal {L}_{2N}\). We define
Clearly, \((X-X_j)(Y-Y_{j'})=0\) on the complement of \(A_{j}\cap B_{j'}\) and \(X-X_{j}\) and \(Y-Y_{j'}\) are positive random variable with super-exponential (uniform in \(j,j'\)) tails (indeed we can just use the upper tail bounds for X and Y). Using the notation \(\Vert X \Vert _p=\mathbbm {E}(|X|^p)^{1/p}\) and the fact that the p-th norm of the of random variables with super-exponential tails can grow at most linearly in p, we know that there exists a constant C such that for all \(j,j'\) and all \(p\ge 1\) \(||X-X_{j}||_{p}, ||Y-Y_{j'}||_{p}\le Cp.\) Using the Hölder inequality we have
where \(p^{-1}+q^{-1}=1\). By the Cauchy-Schwarz inequality
for some new constant \(C>0\). It therefore follows that
for \(p,q \ge 1\) with \(p^{-1}+q^{-1}=1\).
By Lemma 2.12 below, we have
for some \(c>0\). We choose \(p=u\) so that \(1/q=1-\frac{1}{u}\). Therefore, plugging (2.84) in (2.83) it follows that
for some new constant \(c'\). This establishes (2.78) and Theorem 2.1 follows by summing over \((j,j')\). \(\square \)
Lemma 2.12
In the above set-up, for u large enough and \(j+j'\ge u-1\) we have
for some \(c>0\).
Proof
Notice first that arguing as in the proof of Lemma 2.8, if \(j\ge 2^{-2/3}u\) then \({{\,\mathrm{\mathbbm {P}}\,}}(A_{j})\le e^{-\frac{1}{3}u^{3}+cu^2}\) and similarly if \(j'\ge 2^{-2/3}u\) then \({{\,\mathrm{\mathbbm {P}}\,}}(B_{j'})\le e^{-\frac{1}{3}u^{3}+cu^2}\). Therefore it suffices to consider only the cases \(\max \{j,j'\}\le 2^{-2/3}u\). This, together with \(j+j'\ge u-1\) also implies that \(\min \{j,j'\}\ge (1-2^{-2/3})u-1>0\) for u sufficiently large.
Observe now that
By Theorem 2.7 it follows that the first term is upper bounded by \(e^{-\frac{1}{3}u^{3}+cu^2}\) and hence it suffices to show that
for some \(c>0\).
Since the two geodesics do not intersect, we would like to use the BK inequality to get an upper bound. However, we can not do it directly, since the property of being a geodesic depends on all the random variables and not on subsets. We therefore would show something similar for any paths which are of typical length. First, we need to approximate the events \(A_j\) and \(B_{j'}\).
For \(\varepsilon >0\) to be chosen later, let \({\tilde{A}}_{j}\) denote the event that there exists \(k\in \{1,2, \ldots ,\frac{1}{\varepsilon }\}\) such that \(\Gamma _1(2k\varepsilon N)\ge (j-1)(2N)^{2/3}\). Similarly, let \({\tilde{B}}_{j'}\) denote the event that there exists \(k\in \{1,2, \ldots ,\frac{1}{\varepsilon }\}\) such that \(\Gamma _2(2k\varepsilon N)\le (u-j'+1)(2N)^{2/3}\). By choosing \(\varepsilon =\delta u^{-3/2}\) for \(\delta \) sufficiently small but fixed and arguing as in the proof of Theorem 2.3 it follows that
Therefore (2.88) reduces to showing
for some \(c>0\).
Let \(C_2\) be such that (using Lemma A.1)
Observe that on \({\tilde{A}}_j\cap {\tilde{B}}_{j'}\cap \{\Gamma _{1}\cap \Gamma _2 =\emptyset \} \cap \{X>-C_2u\}\cap \{Y>-C_2u\}\) there exist disjoint paths \(\gamma _1\) and \(\gamma _2\) from (0, 0) and I(u) respectively to \(\mathcal {L}_{2N}\) with \(L(\gamma _1), L(\gamma _2)\ge 4N-C_2u 2^{4/3}N^{1/3}\) such that there exist \(k_1,k_2\in \{1,2,\ldots , \frac{1}{\varepsilon }\}\) with \(\gamma _1(2k_1\varepsilon N)\ge (j-1)(2N)^{2/3}\) and \(\gamma _2(2k_2\varepsilon N)\le (u-j'+1)(2N)^{2/3}\). By using the BK inequality as before we get that
where \({\hat{A}}_{j}\) denotes the event that there exists a path \(\gamma _1\) from (0, 0) to \(\mathcal {L}_{2N}\) satisfying \(L(\gamma _1)>4N-C_2u2^{4/3}N^{1/3}\) and \(\gamma _1(2k\varepsilon N)\ge (j-1)(2N)^{2/3}\) for some k and \({\hat{B}}_{j'}\) denotes the event that there exists a path \(\gamma _2\) from I(u) to \(\mathcal {L}_{2N}\) satisfying \(L(\gamma _2)>4N-C_2u2^{4/3}N^{1/3}\) and \(\gamma _2(2k\varepsilon N)\le (u-j-1)(2N)^{2/3}\) for some k. We claim that
Postponing the proof of (2.93) for now, let us first complete the proof of the lemma. By Jensen’s inequality together with \(j+j'\ge u-1\) it follows that
and hence
for some \(c>0\). This, together with (2.92) establishes (2.90).
To conclude the proof we show (2.93). The idea is to follow the proof of Lemma 2.6. However the first bound (2.19) in that proof applies only to geodesics, while here we have to show it for any paths with a length larger that \(4N-C_2 u 2^{4/3}N^{1/3}\). We will prove that for any path \(\gamma _1\) satisfying the conditions of \({{\hat{A}}}_j\), for any \(\tau \in \{\varepsilon ,2\varepsilon ,\ldots ,1\}\)
for \(M=\sqrt{C_2 u+ u^2}\). Then the rest of the proof of Lemma 2.6 applies, except that the sum in (2.24) goes until \(M-u-1\).
Denote \(K(v)=(\tau N,\tau N)+v(2N)^{2/3}(1,-1)\) and divide the possible points where \(\gamma _1\) crosses the line \(\mathcal {L}_{2\tau N}\) as
with \({{\mathcal {I}}}_0=\{K(v),v\in [N^{1/10},\tfrac{\tau N}{(2N)^{2/3}}]\}\) and \({{\mathcal {I}}}_\ell =\{K(v), v\in [\ell ,\ell +1)\}\). Then we have, for any choice of \(A_\ell \) and \(B_\ell =4N-C_2 u 2^{4/3}N^{1/3}-A_\ell \),
By rescaling Lemma A.3, for \(s_1\ll (\tau N)^{2/9}\) and \(u\ll (\tau N)^{1/9}\)
and by rescaling Proposition A.5 (and union bound on \((1-\tau )^{-2/3}\) subsegments per \((2N)^{2/3}\)-length), for \(s_2\ll (1-\tau )^{2/3}N^{2/3}\), we get
We take, with \(\alpha _\ell =-\frac{\ell ^2 (1-\tau )^{1/3}+C_2 u \tau ^{4/3}}{((1-\tau )^{1/3}+\tau ^{1/3})\tau }\), \(A_\ell =4\tau N+\alpha _\ell 2^{4/3}n^{1/3}\). Setting \(\ell =M+{{\tilde{\ell }}}\), we get
From this, it follows that
and thus \(\sum _{\ell =M}^{N^{1/10}-1}(2.102)\le C' \delta ^{-2/3}u^2 e^{-\frac{4}{3} u^3}\), while for \(\ell =0\) it is of an order \(\mathcal {O}(e^{-\frac{4}{3} N^{3/20}})\) smaller. Applying union bound on the \(\varepsilon ^{-1}=u^{3/2}/\delta \) time intervals and the estimate (2.96), we get that any path in \({{\hat{A}}}_j\) is localized within a distance \(M(2N)^{1/3}\) with probability at least \(1-C e^{-u^3}\). \(\square \)
3 Lower Bound
In this section we prove the lower bound of Theorem 1.1.
Theorem 3.1
There exist a constant \(c>0\) such that
We begin explaining the strategy of the proof. Hoeffding’s covariance identity [36], which comes from integration by parts on \(\mathbb {R}_+\) and \(\mathbb {R}_-\) separately, states that
We therefore define the following functions
where in the last equality we used the stationarity of \({{\mathcal {A}}}_1\). As we would like to use (3.2) with \(X={{\mathcal {A}}}_1(0)\) and \(Y={{\mathcal {A}}}_1(u)\), we are interested in finding the asymptotic behavior of
As \(u\rightarrow \infty \), the random variables \({{\mathcal {A}}}_1(0)\) and \({{\mathcal {A}}}_1(u)\) become independent of each other; thus we define \({\mathcal {E}}\) through
where \({{\mathcal {E}}}\rightarrow 0\) when \(u\rightarrow \infty \) (at least for \(s_1\) and \(s_2\) independent of u). Using (3.5) in (3.4) and (3.2) we obtain
Next, by the FKG inequality, see Lemma 3.2 below, the integrand in (3.6) is positive for all \(u\ge 0\). We can therefore restrict the integration in (3.6) to a compact subset of \(\mathbb {R}^2\) to obtain the following lower bound
for any choice of \(\alpha <\beta \).
Thus the goal of the computations below is to show that \({\mathcal {E}}\) is of order \(e^{-\frac{4}{3} u^3}\) times a subleading term for \(s_1,s_2\) in some chosen intervals of size \(\mathcal {O}(1)\) where \(f(s_1)\) and \(f(s_2)\) are bounded away from 0.
As mentioned in the introduction, the strategy taken by Widom in [57], where he obtained the first order terms in the asymptotics of \(F(u;s_1,s_2)\), cannot be fully applied for the \(\hbox {Airy}_1\) process. The strategy can be followed only until Lemma 3.3, to get an expression of \({{\mathcal {E}}}(u;s_1,s_2)\) as a Fredholm determinant of some operator with scalar kernel \({\widetilde{K}}\). It is from here that our proof deviates from the one in [57]. In [57], the author used known identities related to the \(\hbox {Airy}_2\) kernel, to obtain an expression for the trace of \({\widetilde{K}}\), which turns out to be of the simple form \(-\frac{f'(s_1)}{f(s_1)}\frac{f'(s_2)}{f(s_2)}u^{-2}+\mathcal {O}(u^{-4})\). As such identities are not known for the \(\hbox {Airy}_1\) kernel, we compute the trace through the steepest descent method. To compute the covariance one needs to integrate \(f(s_1)f(s_2)\mathcal{E}(u;s_1,s_2)\) over \(\mathbb {R}^2\) (see (3.6)). For the \(\hbox {Airy}_2\) case, one then has \(f(s_1)f(s_2)\mathcal{E}(u;s_1,s_2)=-f'(s_1)f'(s_2) u^{-2}+\mathcal {O}(u^{-4})\), while in the case of the \(\hbox {Airy}_1\) the function \({{\mathcal {E}}}(u;s_1,s_2)\) has no simple asymptotic form. Hence direct computation of the integral seems a difficult task. To overcome this, we restrict the integration in (3.6) to a compact interval. The trick here is to use the FKG inequality applied to the prelimiting LPP model.
Lemma 3.2
For any \(s_1,s_2\in \mathbb {R}\),
Proof
Recalling the notation from (2.4), notice that both \(L^{*}_{N}(0)\) and \(L_{N}^{*}(u)\) are increasing function of the weights \(\omega _{i,j}\sim \exp (1)\) (and they depend only on finitely many vertex weights). For any \(t_1,t_2\) it therefore follows that the events \(\{L^{*}_{N}(0)\le t_1\}\) and \(\{L^{*}_{N}(u)\le t_2\}\) are both decreasing and hence by the FKG inequality they are positively correlated (note that the FKG inequality is often stated for measures on finite distributive lattices satisfying the FKG lattice condition, but more general versions for product measures on finite products of totally ordered measure spaces applicable in the above scenario are available; see e.g. Lemma 2.1 of [42] or Corollary 2 of [41]), and therefore
Using that the \(\hbox {Airy}_1\) process is a scaling limit of \(L^*\) (see (2.5)), the proof is complete. \(\square \)
We first derive an expression for \({{\mathcal {E}}}(u;s_1,s_2)\). Let us begin with the Fredholm representation of the function F. We have from [19, 28, 52]
where K is a \(2\times 2\) matrix kernel
with entries given by the extended kernel of the \(\hbox {Airy}_1\) process [19, 28, 52]
where \({{\,\textrm{Ai}\,}}\) denotes the Airy function. Also, for the one-point distributions, we have \(f(s_i)=\det (\mathbbm {1}-K_{i,i})\) for \(i=1,2\).
The first step is the following result.
Lemma 3.3
With the above notations
where
Proof
Similar to [57], we compute
where we set \(G=-(\mathbbm {1}-K_{1,1})^{-1}K_{1,2}\) and \(H=-(\mathbbm {1}-K_{2,2})^{-1}K_{2,1}\). Moreover,
where
Since \(\det (\mathbbm {1}-K_{\ell ,\ell })=f(s_\ell )\), for \(\ell =1,2\) as we mentioned above, (3.13) follows from (3.5). \(\square \)
Next we would like to approximate the Fredholm determinant in (3.13) by that of a simpler kernel.
Proposition 3.4
Let us define
Then, for any \(s_1,s_2\ge 0\), there exists a constant \(C_2>0\) such that
for any \(u\ge \max \{\frac{1}{2},\sqrt{s_1+s_2}\}\).
The proof of this proposition is in Sect. 3.3.
3.1 The leading term
Lemma 3.3 and Proposition 3.4 suggest
For a trace class operator \({\mathcal {K}}\), the Fredholm determinant is given by
When \({{\mathcal {K}}}=K_{1,2}K_{2,1}\), this translates to
where \({{\,\textrm{tr}\,}}\big (K_{1,2}K_{2,1}\big )=\int _\mathbb {R}dx \int _\mathbb {R}dy K_{1,2}(x,y)K_{2,1}(y,x)\) and
From (3.12), it is clear that the upper tail of \(K_{1,2}K_{2,1}\) in either variables, is determined by that of the function \({{\,\textrm{Ai}\,}}\), which is known to decay super-exponentially, see (A.2). As each of the determinants in (3.23) is a sum of products of elements of the order of \({{\,\textrm{tr}\,}}\big (K_{1,2}K_{2,1}\big )\), one expects the latter to dominate \(R_2\) and therefore that
if \({{\,\textrm{tr}\,}}(K_{1,2}K_{2,1})\) is small.
Let us move on to the computation of \({{\,\textrm{tr}\,}}(K_{1,2}K_{2,1})\). We write the kernel entries \((K_{1,2} K_{2,1})(x,y)\) as well as its trace using complex integral representations, which will then be analyzed. We start with the following identities (see e.g. Appendix A of [6] for the first and last, while the second is a standard Gaussian integral)
where
Plugging these identities into (3.12) we can write
This leads to the following representation of the kernel \(K_{1,2}K_{2,1}\):
Setting \(x=y\) and integrating over x (on \([s_1,\infty )\) due to the indicator functions) we get
We begin with determining the asymptotic behavior of \({{\,\textrm{tr}\,}}(K_{1,2}K_{2,1})\).
Proposition 3.5
For all \(0\le s_1,s_2\le \sqrt{u}\),
as \(u\rightarrow \infty \).
Proof
To get the asymptotic behavior of the trace, we need to choose the parameters a, b in the integration contours. We do it in a way that the dominant parts in the exponents of the contour integrals in (3.30) are minimized.
Now we need to choose the parameters a, b. For that reason we search for the minimizers of the different exponents in (3.32):
-
(a)
for
$$\begin{aligned} \frac{d}{d\xi _1}\Big (-\frac{\xi _1^3}{3}+u\xi _1^2+(s_1+s_2)\xi _1\Big )=2u\xi _1-\xi _1^2=0, \end{aligned}$$(3.33)which is solved for \(\xi _1=u-\sqrt{u^2+s_1+s_2}=:a_1\) or \(\xi _1=u+\sqrt{u^2+s_1+s_2}\). The solution which satisfy the constraint \({\text {Re}}(\xi _1)<u\) in (3.26) is also the minimum.
-
(b)
For
$$\begin{aligned} \frac{d}{d\xi _2}\Big (u\xi _2^2+(s_1-s_2)\xi _2\Big )=2u\xi _2+(s_1-s_2) \end{aligned}$$(3.34)we see that \(\xi _2=(s_2-s_1)/(2u)=:a_2\) is the minimum.
-
(c)
Similarly,
$$\begin{aligned} \frac{d}{d\eta }\Big (\frac{\eta ^3}{3}-u\eta ^2-(s_1+s_2)\eta \Big )=\eta ^2-2u\eta -(s_1+s_2)=0, \end{aligned}$$(3.35)has the minimum is at \(\eta =u+\sqrt{u^2+s_1+s_2}=:b\) satisfies \({{\textrm{Re}}(\eta )}>u\).
So let us use the following change of variables
with \(z,w\in \textrm{i}\mathbb {R}\) into (3.32). The two terms are analyzed in the same way, thus we write the details only for the first one.
Denote \(\sigma =(s_1+s_2)/u^2\le 2 u^{-3/2}\) and consider the first term in (3.32). We have
and for the prefactorFootnote 4
Each term in the exponential which is cubic in \(z,w\in \textrm{i}\mathbb {R}\) is purely imaginary, thus its exponential is bounded by 1. Furthermore, the quadratic terms in z and w have a positive prefactor \(\sqrt{1+\sigma }\ge 1\). Thus integrating over \(|z|>u^{1/4}\) and/or \(|w|>u^{1/4}\) we get a correction term of order \(e^{-\sqrt{u}}\) times the value of the integrand at \(z=w=0\). For the rest of the integral, for which \(|z|,|w|\le u^{1/4}\), the error terms \(\mathcal {O}(z u^{-3/2};w u^{-3/2};\sigma )=\mathcal {O}(u^{-5/4})\).
So the first term in (3.32) is given by
Finally, extending the Gaussian integrals to \(\textrm{i}\mathbb {R}\), we get an error term \(\mathcal {O}(e^{-\sqrt{u}})\) only and using
we get
A similar computation for the second term in (3.32) leads to
Summing up (3.41) and (3.42) we obtain
\(\square \)
3.2 Bounding lower order terms
To show (3.24) we need to get a bound on \(R_2(u;s_1,s_2)\) from (3.23), which is \(o({{\,\textrm{tr}\,}}\big (K_{1,2}K_{2,1}\big ))\).
Proposition 3.6
There exists a constant \(C>0\) such that
uniformly in \(s_1,s_2\in \mathbb {R}\).
Proof
In (3.29) we use the change of variables
where \(z,w\in \textrm{i}\mathbb {R}\). This leads to
with
Since \(x,y\in \mathbb {R}\) and \(z,w\in \textrm{i}\mathbb {R}\), we get the simple bound \(|P(z,w,x,y)|\le \frac{1}{u}\), while \(e^{z^2+w^2}\) is real. Performing the Gaussian integral we then get
Let \(K=K_{1,2}K_{2,1}\). Hadamard’s inequalityFootnote 5 gives
so that
Denote \(M=\frac{e^{-\frac{{4u^3}}{3}-2(s_1+s_2)u}}{8\pi u^3}\). Then there exists \(C>0\) such that
This completes the proof. \(\square \)
We have now all the ingredients to complete the proof of Theorem 3.1.
Proof of Theorem 3.1
We have
where
It follows that
We shall apply the lower bound (3.7) for the covariance for an appropriate choice of \(\alpha =\alpha (u)\) and \(\beta =\beta (u)\) such the contribution of error term \(f(s_1)f(s_2)R(u;s_1,s_2)\) is of a subleading order.
We shall choose \(\alpha >0\) depending on u and for concreteness set \(\beta =\alpha +1\). Thus we integrate over a region of area 1. By Proposition 3.6, the error term \(R_2(u;s_1,s_2)\) is much smaller than the leading term coming from the trace, see Proposition 3.5 (of order \(e^{-\frac{4}{3} u^3}\) smaller) for \(s_1,s_2\ll u^2\). However, the exponential term in u in the error bound of \(R_1(u;s_1,s_2)\) coming from (3.19) is of the same order, namely \(e^{-2(s_1+s_2)u-\frac{4}{3} u^3}\). The difference is that in the trace we have a prefactor \(\sim \frac{s_1 s_2}{u^4}\), while in the bound for \(R_1\) we have a prefactor \(\sim \frac{e^{-\min \{s_1,s_2\}}}{u^2}\). Thus, in order to ensure that the contribution from \(R_1\) is subleading, we can take \(s_1,s_2\sim c\ln (u)\) for any \(c\ge 3\). Therefore choosing \(\alpha =3\ln (u)\) and using the fact that for all \(x\ge 0\), \((x)=F_{\textrm{GOE}}(2^{2/3}x)\) is bounded away from 0 (in fact, one can numerically check \(f(x)=F_{\textrm{GOE}}(2^{2/3}x)\in [F_\textrm{GOE}(0),1]=[0.83...,1]\)), we finally get the claimed result. \(\square \)
3.3 Proof of Proposition 3.4
To prove Proposition 3.4, we begin with the following well known bound (see e.g. Lemma 4(d), Chapter XIII.17 of [51])
where \(Q={\tilde{K}}-K_{1,2}K_{2,1}\), where \(\Vert \cdot \Vert _1\) is the trace-norm (see e.g. [55]). From Lemmas 3.10 and 3.11 below, the exponent in the display above is bounded and the result follows. \(\square \)
In the remainder of this section we prove the two results Lemma 3.10 and Lemma 3.11 used above. We first need to prove some auxiliary bounds. From the identity
we see that
Recall that from (3.55) we need to bound \(\Vert Q\Vert _1\). Thus it is enough to bound the \(\Vert \cdot \Vert _1\)-norm of each of the terms on the right hand side of (3.57). Since \(K_{1,2}(x,y)\) is not decaying as \(x=y\rightarrow \infty \), we could either work in weighted \(L^2\) spaces, or as we do here, introduce some weighting in the kernel elements. Namely define
Also, we use the norm inequalities \(\Vert A B\Vert _1\le \Vert A\Vert _2 \Vert B\Vert _2\) and \(\Vert A\Vert _2\le \Vert A\Vert _1\), with \(\Vert \cdot \Vert _2\) the Hilbert-Schmidt norm given by \(\Vert K\Vert _2=\Big (\int dxdy (K(x,y))^2\Big )^{1/2}\).
These norm inequalities, the fact that \(K_{i,i}\) and \((\mathbbm {1}-K_{i,i})^{-1}\) commute and the identity \(K_{1,1}K_{1,2}={\bar{K}}_{1,1}{\bar{K}}_{1,2}^L\) lead to
Moreover, using \(K_{1,2}K_{2,2}={\bar{K}}^R_{1,2}{\bar{K}}_{2,2}\), we get
and finally
We turn to bound each of the terms on the right hand side of each of the inequalities in (3.59)–(3.61).
We also use the following change of variables: for \(a,b\in \mathbb {R}\), define
which give \(x=a+\frac{1}{2}(\omega +\zeta )\) and \(y=b+\frac{1}{2}(\omega -\zeta )\). So the integral over \((x,y)\in [a,\infty )\times [b,\infty )\) becomes an integral over \((\omega ,\zeta )\) with \(\omega \ge 0\) and \(\zeta \in [-\omega ,\omega ]\).
Lemma 3.7
Uniformly for all \(s_1,s_2\ge 0\),
for \(i\in \{1,2\}\).
Proof
The bounds for \(i=1\) and \(i=2\) are fully analogous, thus we restrict to the case \(i=1\) here. To bound \(\Vert K_{1,1}\Vert _2^2\) we use (A.3) which gives
Thus \(\Vert K_{1,1}\Vert _2\le \frac{1}{2} e^{-2 s_1}\). Similarly
which implies the second bound in (3.63). Finally using
which holds whenever \(\Vert K_{i,i}\Vert _2<1\), we get the last inequality. \(\square \)
Lemma 3.8
For \(s_1,s_2\ge 0\) and \(u>0\)
Proof
Using the change of variables (3.62) with \(a=s_2,b=s_1\), it follows that
where we used that the Jacobian of the transformation in (3.62) is 2.
Next using (A.5) (with \(x=\omega +s_1+s_2\)) we get
\(\square \)
Lemma 3.9
For \(s_1,s_2\ge 0\), there exists a constant \(C_1>0\) such that
for all \(u\ge \sqrt{s_1+s_2}\).
Proof of Lemma 3.9
By symmetry in x, y the bounds for \(\Vert {\bar{K}}^L_{1,2}\Vert _2^2\) and for \(\Vert {\bar{K}}^R_{1,2}\Vert _2^2\) are the same. So we consider \(\Vert {\bar{K}}^L_{1,2}\Vert _2^2\) only. Using \((A-B)^2\le 2 (A^2+B^2)\) we get
The second term in (3.71) can be bounded by integrating in y over \(\mathbb {R}\) and then integrating over \(x\ge s_2\). This gives
For the first term in (3.71), we use the change of variable (3.62) with \(a=s_1\), \(b=s_2\) and obtain
Let \(A(\omega )=-\frac{4}{3} (\omega +s_1+s_2+u^2)^{3/2}+2 (\omega +s_1+s_2) u+\tfrac{4}{3} u^3\). Then for all \(s_1,s_2,u\ge 0\) it is a concave function in \(\omega \) and thus
The term independent of \(\omega \) is always negative. Indeed, \(x\mapsto \frac{4}{3}(x+u^2)^{3/2}\) is convex and thus greater than its linear approximation at \(x=0\), which is \(\frac{4}{3} u^3+2ux\). So we have
where in the last step we used the assumption that \(u\ge \sqrt{s_1+s_2}\). \(\square \)
Lemma 3.10
There exists a constant \(C_1>0\) such that for \(s_1,s_2\ge 0\),
for all \(u\ge \max \{\frac{1}{2},\sqrt{s_1+s_2}\}\).
Proof
Denote
Then we have the bound
From Lemma 3.9, we get \(\Vert {\bar{K}}_{1,2}^{R}\Vert _2\le \frac{C_1}{u^{1/2}}\). Furthermore, using (A.5), we have
for all \(u\ge 1/2\). \(\square \)
We are now ready to bound \(\Vert Q\Vert _1\).
Lemma 3.11
Uniformly for \(s_1,s_2\ge 0\), there exists a constant \(C_1\) such that
for all \(u\ge \max \{\frac{1}{2},\sqrt{s_1+s_2}\}\).
Proof
Applying the bounds of Lemmas 3.7 and 3.9 to (3.59)–(3.61), we obtain
for all \(s_1,s_2\ge 0\) and u as mentioned. Then, applying the bound of Lemma 3.8 leads to
\(\square \)
Notes
To be precise, in the mentioned papers, they obtained explicit expressions for the joint distribution of \({{\mathcal {A}}}_2(0)\) and \({{\mathcal {A}}}_2(u)\). To get the covariance one needs to integrate it and each of the terms in inverse powers of u can be explicitly integrated. A rigorous control of the error term is not carried out, but in [54] a numerical comparison up to \(u^{-10}\) is provided confirming the correctness.
We do not write explicitly the rounding to integer values, i.e., (x, y) stands for \((\lfloor x\rfloor ,\lfloor y\rfloor )\).
The tail bounds in Corollary A.2 clearly continue to hold even after removing the random variable at K(v) which is \(\textrm{Exp}(1)\)-distributed. The advantage is that in this way \(L_{(0,0),K(v)}\) and \(\tilde{L}_{K(v),\mathcal {L}_{2N}}\) are independent random variables.
The notation \(\mathcal {O}(a_1;...;a_k)\) stands for \(\mathcal {O}(a_1)+...+\mathcal {O}(a_k)\).
Let A be a \(n\times n\) matrix with \(|A_{i,j}|\le 1\). Then \(|\det (A)|\le n^{n/2}\).
\(\sup _{x\in \mathbb {R}}|{{\,\textrm{Ai}\,}}(x)|\le c=0.7857\ldots \) follows by \(\lim _{n\rightarrow \infty } n^{1/3} J_{[2n+u n^{1/3} u]}(2n)={{\,\textrm{Ai}\,}}(u)\) (see also (3.2.23) of [1]) and the bound of Landau [43]. For any \(x\ge {0.01}\), the bound (A.4), is better that the bound in (A.3) and \(e^{-0.01}>c\).
Removing the end-point does not influence the asymptotics and bounds, but it has the property that \({{\hat{L}}}_{w,I(u)}\) and \(L_{I(u),\mathcal {L}_{2N}}\) are independent random variables and the concatenation property holds true.
References
Abramowitz, M., Stegun, I.A.: Pocketbook of Mathematical Functions. Verlag Harri Deutsch, Thun-Frankfurt a. M. (1984)
Adler, M., van Moerbeke, P.: PDE’s for the joint distribution of the Dyson, Airy and Sine processes. Ann. Probab. 33, 1326–1361 (2005)
Arratia, R., Garibaldi, S., Hales, A.W.: The van den Berg–Kesten–Reimer operator and inequality for infinite spaces. Bernoulli 24(1), 433–448 (2018)
Baik, J.: Painlevé expressions for LOE, LSE and interpolating ensembles. Int. Math. Res. Not. 1739–1789, 2002 (2002)
Baik, J., Buckingham, R., DiFranco, J.: Asymptotics of Tracy–Widom distributions and the total integral of a Painleve II function. Commun. Math. Phys. 280, 463–497 (2008)
Baik, J., Ferrari, P.L., Péché, S.: Limit process of stationary TASEP near the characteristic line. Commun. Pure Appl. Math. 63, 1017–1070 (2010)
Baik, J., Liechty, K., Schehr, G.: On the joint distribution of the maximum and its position of the Airy\(_2\) process minus a parabola. J. Math. Phys. 53, 083303 (2012)
Baik, J., Rains, E.M.: Algebraic aspects of increasing subsequences. Duke Math. J. 109, 1–65 (2001)
Baik, J., Rains, E.M.: The asymptotics of monotone subsequences of involutions. Duke Math. J. 109, 205–281 (2001)
Barraquand, G., Krajenbrink, A., Le Doussal, P.: Half-space stationary Kardar–Parisi–Zhang equation beyond the Brownian case (2022). arXiv:2202.10487
Basu, R., Ganguly, S.: Time correlation exponents in last passage percolation. In: Vares, M.E., Fernández, R., Fontes, L.R., Newman, C.M. (eds.) In and Out of Equilibrium 3: Celebrating Vladas Sidoravicius, volume 77 of Progress in Probability. Birkhäuser, Berlin (2021)
Basu, R., Ganguly, S., Zhang, L.: Temporal correlation in last passage percolation with flat initial condition via Brownian comparison. Commun. Math. Phys. 383(3), 1805–1888 (2021)
Basu, R., Sarkar, S., Sly, A.: Coalescence of geodesics in exactly solvable models of last passage percolation. J. Math. Phys. 60(9), 093301 (2019)
Basu, R., Sidoravicius, V., Sly, A.: Last passage percolation with a defect line and the solution of the slow bond problem (2014). arXiv:1408.3464
Bornemann, F.: On the numerical evaluation of Fredholm determinants. Math. Comput. 79, 871–915 (2009)
Bornemann, F., Ferrari, P.L., Prähofer, M.: The Airy\(_1\) process is not the limit of the largest eigenvalue in GOE matrix diffusion. J. Stat. Phys. 133, 405–415 (2008)
Borodin, A., Ferrari, P.L.: Large time asymptotics of growth models on space-like paths I: PushASEP. Electron. J. Probab. 13, 1380–1418 (2008)
Borodin, A., Ferrari, P.L., Prähofer, M.: Fluctuations in the discrete TASEP with periodic initial configurations and the Airy\(_1\) process. Int. Math. Res. Pap. 2007, rpm002 (2007)
Borodin, A., Ferrari, P.L., Prähofer, M., Sasamoto, T.: Fluctuation properties of the TASEP with periodic initial configuration. J. Stat. Phys. 129, 1055–1080 (2007)
Borodin, A., Ferrari, P.L., Sasamoto, T.: Transition between Airy\(_1\) and Airy\(_2\) processes and TASEP fluctuations. Commun. Pure Appl. Math. 61, 1603–1629 (2008)
Borodin, A., Gorin, V.: Lectures on integrable probability. In: Probability and Statistical Physics in St. Petersburg. Proceedings of Symposia in Pure Mathematics, vol. 91, pp. 155–214. AMS (2016)
Borodin, A., Péché, S.: Airy Kernel with two sets of parameters in directed percolation and random matrix theory. J. Stat. Phys. 132, 275–290 (2008)
Busani, O., Ferrari, P.L.: Universality of the geodesic tree in last passage percolation. Ann. Probab. 50, 90–130 (2022)
Chhita, S., Ferrari, P.L., Spohn, H.: Limit distributions for KPZ growth models with spatially homogeneous random initial conditions. Ann. Appl. Probab. 28, 1573–1603 (2018)
Corwin, I.: The Kardar–Parisi–Zhang equation and universality class. Random Matrices Theory Appl. 01, 1130001 (2012)
Corwin, I., Ferrari, P.L., Péché, S.: Universality of slow decorrelation in KPZ models. Ann. Inst. H. Poincaré Probab. Stat. 48, 134–150 (2012)
Dimitrov, E.: Two-point convergence of the stochastic six-vertex model to the Airy process (2020). arXiv:2006.15934
Ferrari, P.L.: The universal Airy\(_1\) and Airy\(_2\) processes in the Totally Asymmetric Simple Exclusion Process. In: J. Baik, T. Kriecherbauer, L.-C. Li, K. McLaughlin, C. Tomei (eds.) Integrable Systems and Random Matrices: In Honor of Percy Deift, Contemporary Mathematics, pp. 321–332. Amer. Math. Soc., Providence (2008)
Ferrari, P.L.: From interacting particle systems to random matrices. J. Stat. Mech. 2010, P10016 (2010)
Ferrari, P.L., Occelli, A.: Universality of the GOE Tracy–Widom distribution for TASEP with arbitrary particle density. Eletron. J. Probab. 23(51), 1–24 (2018)
Ferrari, P.L., Occelli, A.: Time-time covariance for last passage percolation with generic initial profile. Math. Phys. Anal. Geom. 22, 1 (2019)
Ferrari, P.L., Occelli, A.: Time-time covariance for last passage percolation in half-space (2022). arXiv:2204.06782
Ferrari, P.L., Spohn, H.: Random growth models. In: Akemann, G., Baik, J., Di Francesco, P. (eds.) The Oxford Handbook of Random Matrix Theory, pp. 782–801. Oxford Univ. Press, Oxford (2011)
Ferrari, P.L., Spohn, H.: On time correlations for KPZ growth in one dimension. SIGMA 12, 074 (2016)
Hammond, A., Sarkar, S.: Modulus of continuity for polymer fluctuations and weight profiles in Poissonian last passage percolation. Electron. J. Probab. 25, 38 (2020)
Hoeffding, W.: Masstabinvariante Korrelationstheorie. Schriften Math. Inst. Univ. Berlin 5, 181–233 (1940)
Johansson, K.: Shape fluctuations and random matrices. Commun. Math. Phys. 209, 437–476 (2000)
Johansson, K.: Discrete polynuclear growth and determinantal processes. Commun. Math. Phys. 242, 277–329 (2003)
Johansson, K.: The arctic circle boundary and the Airy process. Ann. Probab. 33, 1–30 (2005)
Kardar, M., Parisi, G., Zhang, Y.Z.: Dynamic scaling of growing interfaces. Phys. Rev. Lett. 56, 889–892 (1986)
Kemperman, J.H.B.: On the FKG-inequality for measures on a partially ordered space. In: Indagationes Mathematicae (Proceedings), vol. 80, no. (4), pp. 313–331. North-Holland (1977)
Kesten, H.: First-passage percolation. In: Picco, P., San Martin, J. (eds.) From Classical to Modern Probability: CIMPA Summer School 2001, pp. 93–143. Birkhäuser Basel, Basel (2003)
Landau, L.J.: Bessel functions: monotonicity and bounds. J. Lond. Math. Soc. 61, 197–215 (2000)
Ledoux, M., Rider, B.: Small deviations for beta ensembles. Electron. J. Probab. 15, 1319–1343 (2010)
NIST Digital Library of Mathematical Functions. Release 1.1.5 of 2022-03-15. F.W.J. Olver, A.B. Olde Daalhuis, D.W. Lozier, B.I. Schneider, R.F. Boisvert, C.W. Clark, B.R. Miller, B.V. Saunders, H.S. Cohl, M.A. McClain (eds.) http://dlmf.nist.gov/,
Pimentel, L.P.R.: Duality between coalescence times and exit points in last-passage percolation models. Ann. Probab. 44(5), 3187–3206 (2016)
Pimentel, L.P.R.: Local Behavior of Airy Processes. J. Stat. Phys. 173, 1614–1638 (2018)
Prähofer, M., Spohn, H.: Scale invariance of the PNG droplet and the Airy process. J. Stat. Phys. 108, 1071–1106 (2002)
Quastel, J.: Introduction to KPZ. In: Current developments in mathematics, pp. 125–194. International Press, Somerville, MA (2012)
Quastel, J., Spohn, H.: The one-dimensional kpz equation and its universality class. J. Stat. Phys. 160, 965–984 (2015)
Reed, M., Simon, B.: Methods of Modern Mathematical Physics III: Scattering theory. Academic Press, New York (1978)
Sasamoto, T.: Spatial correlations of the 1D KPZ surface on a flat substrate. J. Phys. A 38, L549–L556 (2005)
Schehr, G.: Extremes of \(N\) vicious walkers for large \(N\): application to the directed polymer and KPZ interfaces. J. Stat. Phys. 149, 385–410 (2012)
Shinault, G., Tracy, C.: Asymptotics for the covariance of the Airy\(_2\) process. J. Stat. Phys. 143, 60–71 (2011)
Simon, B.: Trace Ideals and their Applications, 2nd edn. American Mathematical Society, Providence (2000)
Takeuchi, K.A.: An appetizer to modern developments on the Kardar–Parisi–Zhang universality class. Phys. A 504, 77–105 (2016)
Widom, H.: On asymptotic for the Airy process. J. Stat. Phys. 115, 1129–1134 (2004)
Zhang, L.: Optimal exponent for coalescence of finite geodesics in exponential last passage percolation. Electron. Commun. Probab. 25, 14 (2020)
Zygouras, N.: Some algebraic structures in the KPZ universality (2018). arXiv:1812.07204
Acknowledgements
The work of O. Busani and P.L. Ferrari was partly funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy - GZ 2047/1, projekt-id 390685813. P.L. Ferrari was also supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - Projektnummer 211504053 - SFB 1060. R. Basu is partially supported by a Ramanujan Fellowship (SB/S2/RJN-097/2017) and a MATRICS grant (MTR/2021/000093) from SERB, Govt. of India, DAE project no. RTI4001 via ICTS, and the Infosys Foundation via the Infosys-Chandrasekharan Virtual Centre for Random Geometry of TIFR.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interests
The authors have no relevant financial or non-financial interests to disclose.
Additional information
Communicated by K. Johansson.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
A Some Tail Estimates
A Some Tail Estimates
Here we collect and, in some cases, prove estimates on tails of the Airy function and tails of some LPPs (point-to-point and point-to-half line) that we have used earlier.
1.1 A.1 Bounds on Airy functions
The Airy function \({{\,\textrm{Ai}\,}}\) satisfied the following identity (see e.g. Appendix A of [6])
for any \(\gamma =\alpha +\textrm{i}\mathbb {R}\) with \(\alpha >-b\). The asymptotics of \({{\,\textrm{Ai}\,}}\) is given by [1]
Furthermore, one has the following simple boundsFootnote 6
and, see Equation 9.7.15 of [45],
from which it follows that for all \(x,u\ge 0\),
using that \(\frac{2}{3}(x+u^2)^{3/2}\ge \frac{2}{3} u^3+ux\) since \(y\mapsto y^{3/2}\) is convex on the positive real line.
1.2 A.2 Bounds on LPP
Lemma A.1
(Theorem 2 of [44]). There exist constants \(C,c>0\) such that
for all \(N^{2/3}\gg x>0\) and \(N^{1/3}\gg |v|\).
Using Riemann–Hilbert methods like as in the case of geometric and Poissonian LPP, see [4, 5, 8, 9] for the Riemann–Hilbert problem for exponential LPP, it should be possible to get the optimal constant c as well (expected to be \(\frac{1}{12}\) like for the lower tail of the GUE Tracy–Widom distribution function), but we do not require the optimal constant in the lower tail estimate for our purposes. A simple corollary is the following.
The following corollary follows from the inequality \(L_{(0,0),\mathcal {L}_{2N}}\ge L_{(0,0),(N,N)}\).
Corollary A.2
There exist constants \(C,c>0\) such that
for all \(N^{2/3}\gg x>0\).
Next we would like to have a sharp upper tail moderate deviation estimate from the last passage time from (0, 0) to the half line \(\{(N,N)+v(2N)^{2/3}(1,-1):v\ge u\}\).
Lemma A.3
Let \({{\mathcal {D}}}_u=\cup _{v\ge u} J(v)\) with \(J(v)=(N,N)+v(2N)^{2/3}(1,-1)\). Then, for \(N^{1/9}\gg u>0\) and \(N^{2/9}\gg s>0\),
for some constant C.
Proof
Let \({{\tilde{\mathcal {L}}}}=\{(x,y)\in \mathbb {Z}^2| x+y=0, y\ge 0\}\). Then for any a,
Next we use the well-known connection between TASEP and LPP (see e.g. Equation (1.15) of [6], which generalizes [37]). In our case, the connection with TASEP with half-flat initial condition, i.e., \(x_k(0)=-2k\) for \(k\ge 0\) are the positions of TASEP particles at time 0. This gives
with \(m=N-u (2N)^{2/3}\) and the sum is on \(x_1,\ldots ,x_n\) being integers below \(2u(2N)^{2/3}\); the kernel is given by [20]
with \(\Phi (v;m,y)=-v t+(y+m)\ln (1+v)-m\ln (v)\). The contours are simple paths anticlockwise oriented, with \(\Sigma _0\) enclosing only the pole at 0, while \(\Sigma _{-1,-1-v}\) enclosing the poles at \(-1\) and \(-1-v\). Consider the scaling
Then, with \(\delta =2^{-4/3}N^{-1/3}\), we have
The reason why we singled out the \(\delta ^n\) is because one can think that when \(\delta \) is small, the Riemann sum \(\sum _{\xi _1,\ldots ,\xi _n\in \delta \mathbb {N}} \delta ^n f(\xi _1,\ldots ,\xi _n)\) is close to the integral \(\int _{\mathbb {R}_+^n} d\xi _1\ldots d\xi _n f(\xi _1,\ldots ,\xi _n)\). Actually, from the exponential bound we get for \(f(\xi _1,\ldots ,\xi _n)\) we can bound the estimate with the integral (starting from \(-\delta \) instead of 0 to be precise).
Now we give the choice of the paths for v and w (which are such that they are steep descent). Since it is quite standard, we just indicate some main steps (see e.g. [20] or Lemma 2.7 of [24] for a case very similar to the one in this paper). Choose the paths as
On the chosen paths, we have, for \(u\ll N^{1/3}\) and \(s\ll N^{2/3}\)
Furthermore
as the maximal value of the first expression is obtained for the values \(\theta =\phi =0\), i.e., when v and w are closest to each other, while \(1+w\) and v are two circles around 0 with radius difference \(\varepsilon _1-\varepsilon _2\). Putting all together and noticing that \(\int _{-\pi }^\pi d\theta e^{-c_1 (1-\cos (\theta ))}\le C/\sqrt{c_1}\) we finally obtain
Taylor expansion gives
for all N large enough and for some explicit conjugation terms \(g(\xi )\) (which cancel out exactly when computing the determinant), where \(E_1=\mathcal {O}(s N^{-2/3};u N^{-1/3})=o(1)\) and \(s^{3/2}E_2=\mathcal {O}(s^{5/2} N^{-2/3}; s^{1/2}u^4 N^{-2/3})=o(1)\) do not depend on \(\xi _i,\xi _j\).
Inserting (A.18) and (A.17) in (A.13),using the Hadamard bound and interchanging the order of summation, we get
for some constants \(C',C''\). To bound the sum over \(\xi _i\), we simply used \(\sum _{\xi \in \delta \mathbb {N}}\delta e^{-\alpha \xi }\le \int _0^\infty e^{-\alpha \xi } d\xi =\alpha ^{-1}\), but one could also compute the geometric sum explicitly. \(\square \)
The next result gives upper tail bounds for point-to-line LPP.
Lemma A.4
For \(N^{2/9}\gg s\ge 0\),
for some constant C.
Proof
We need only to prove the result for \(s\ge 1\) (or any other constant). Then by appropriate choice of the constant C, the result holds for all \(s\ge 0\) (just take C such that the upper bound is larger than the trivial bound 1).
We have
where \(x_n(t)\) it the position of TASEP particle n at time t with initial condition \(x_n(0)=-2n\), \(n\in \mathbb {Z}\). The distribution of TASEP particles are given in terms of a Fredholm determinant
with the kernel given by [19]
with \(\Sigma _0\) a simple path anticlockwise oriented enclosing only the pole at 0. Setting \(x_i=-\xi _i 2^{4/3}N^{1/3}\) and \(t=4N+s 2^{4/3}N^{1/3}\) we obtain
with
Consider the path parameterized by \(v=-\rho e^{\textrm{i}\theta }\), \(\theta \in [-\pi ,\pi )\). Then due to
the path is steep descent for any \(\rho \in (0,1/2]\). Let us choose the radius by
For any \(s,\xi _i,\xi _j\ge 0\),
and for \(0<s\ll N^{2/3}\),
Finally, integrating over \(\theta \) leads to the following estimate on the kernel
A computation gives
for some conjugation function g, where the error terms do not depend on \(\xi _i,\xi _j\). Thus we have the following estimate on the kernel
for some constant C.
Plugging this estimate into (A.22) and using Hadamard inequality like in the proof of Lemma A.3 we finally get the claimed bound (A.20). \(\square \)
The final result we need is an upper tail bound for interval-to-line LPP. The proof is similar to that of Lemma 10.6 in [14], but with the optimal exponent.
Proposition A.5
Let \({{\mathcal {M}}}=\{(k,-k)|-\tfrac{1}{2}(2N)^{2/3}\le k\le \tfrac{1}{2}(2N)^{2/3}\}\) and \(\mathcal {L}_{2N}=\{(N+k,N-k),k\in \mathbb {Z}\}\). Then, for \(N^{2/3}\gg s\ge 0\),
for some constant \(C>0\).
Proof
Notice that it is enough to prove the bound only for \(s\ge 1\) (or any other positive constant).
Recall the notation \(I(v)=v(2N)^{2/3}(1,-1)\). So \({{\mathcal {M}}}\) is the union of points I(v) with \(|v|\le \tfrac{1}{2}\). Define the point w
Let us divide \({\mathcal {M}}\) into union of \({{\tilde{\varepsilon }}}^{-2/3}\) segments of size \({{\tilde{\varepsilon }}}^{2/3}(2N)^{2/3}\), say \(\mathcal{M}=\bigcup _{k=1}^{{{\tilde{\varepsilon }}}^{-2/3}} {{\mathcal {M}}}_k\). Then
Using union bound the the fact that each of the \(\sup _{p\in \mathcal{M}_k} L_{p,\mathcal {L}_{2N}}\) has the same distribution, we get
Next we want to bound the last probability in (A.36).
The idea is the following. For fixed v, we know that the same tail estimates we want to prove hold for \(L_{I(v),\mathcal {L}_{2N}}\). We also have an estimate of the upper tail for \(L_{w,\mathcal {L}_{2N}}\) and that with positive probability, \(L_{w,I(v)}\) can not be too small in the \(N^{1/3}\) scale. So, the fluctuations coming from maximizing \(L_{I(v),\mathcal {L}_{2N}}\) over \(|v|\le \tfrac{1}{2}(1-{{\tilde{\varepsilon }}})^{2/3}\) can not be compensated fully from the fluctuations of \(L_{w,I(v)}\). This will imply our claim.
Let \({{\hat{L}}}_{w,I(u)}=L_{w,I(u)}-\omega _{I(u)}\) be the LPP from w to I(u) without the random variable at the end-point I(u) Footnote 7. We know from [30, 47] that
is tight in the space of continuous functions on compact sets. As a consequence there exists a constant \(C'>0\) such that the event
satisfies \({{\,\mathrm{\mathbbm {P}}\,}}({{\mathcal {H}}})\ge 1/2\).
We also have
Define the event
Then,
where we used independence of \({\mathcal {H}}\) and \({\mathcal {G}}\). Thus we have shown that
where the last inequality holds for \(N^{2/3}\gg s-C'{{\tilde{\varepsilon }}}^{1/3}>1\) by Lemma A.4. Replacing this into (A.36) we get
Consider first \(s\ge 1\). Then taking \({{\tilde{\varepsilon }}}=1/s^{3/2}\) (notice that \({{\tilde{\varepsilon }}} N \gg 1\) since we assumed \(s\ll N^{2/3}\)), we get
for some new constant \(C>0\). \(\square \)
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Basu, R., Busani, O. & Ferrari, P.L. On the Exponent Governing the Correlation Decay of the \(\hbox {Airy}_1\) Process. Commun. Math. Phys. 398, 1171–1211 (2023). https://doi.org/10.1007/s00220-022-04544-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00220-022-04544-1