Skip to main content
Log in

Internally connected graphs and the Kashiwara-Vergne Lie algebra

  • Published:
Letters in Mathematical Physics Aims and scope Submit manuscript

Abstract

It is conjectured that the Kashiwara-Vergne Lie algebra \(\widehat{\mathfrak {krv}}_2\) is isomorphic to the direct sum of the Grothendieck-Teichmüller Lie algebra \(\mathfrak {grt}_1\) and a one-dimensional Lie algebra. In this paper, we use the graph complex of internally connected graphs to define a nested sequence of Lie subalgebras of \(\widehat{\mathfrak {krv}}_2\) whose intersection is \(\mathfrak {grt}_1\), thus giving a way to interpolate between these two Lie algebras.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

References

  1. Alekseev, A., Torossian, C.: The Kashiwara-Vergne conjecture and Drinfeld’s associators. Ann. Math. 172(2), 415–463 (2012)

    Article  MathSciNet  MATH  Google Scholar 

  2. Conant, J., Gerlits, F., Vogtmann, K.: Cut vertices in commutative graphs. Q. J. Math. 56(3), 321–336 (2005)

    Article  MathSciNet  MATH  Google Scholar 

  3. Dolgushev, V.A., Rogers, C.L.: Notes on algebraic operads, graph complexes, and Willwacher’s construction. In: Evens, S., Gekhtman, M., Hall, B.C., Liu, X., Polini, C. (eds.) Mathematical Aspects of Quantization: Center for Mathematics at Notre Dame, Summer School and Conference, Notre Dame University, Notre Dame, Indiana, May 31–June 10 2011. Contemporary Mathematics, vol. 583. American Mathematical Society, Providence (2011)

  4. Drinfeld, V.G.: On quasitriangular quasi-Hopf algebras and on a group that is closely connected with \({\rm Gal}(\overline{{\bf Q}}/{\bf Q})\). Leningrad Math. J. 2(4), 829–860 (1991)

    MathSciNet  Google Scholar 

  5. Kashiwara, M., Vergne, M.: The Campbell-Hausdorff formula and invariant hyperfunctions. Invent. Math. 47(3), 249–272 (1978)

    Article  MathSciNet  MATH  ADS  Google Scholar 

  6. Kontsevich, M.: Formal (non)commutative symplectic geometry. The Gel\(\prime \)fand Mathematical Seminars, 1990–1992, pp. 173–187. Birkhäuser Boston, Boston (1993)

    Google Scholar 

  7. Kontsevich, M.: Feynman diagrams and low-dimensional topology. In: Joseph, A., Mignot, F., Murat, F., Prum, B., Rentschler, R. (eds.) First European Congress of Mathematics, Paris, 6–10 July 1992. Progress in Mathematics, vol. 120. Birkhäuser, Basel (1994)

  8. Kontsevich, M.: Formality conjecture. In: Deformation theory and symplectic geometry (Ascona, 1996), volume 20 of Math. Phys. Stud. pp. 139–156. Kluwer Acadmic Publication, Dordrecht (1997)

  9. Kontsevich, M.: Operads and motives in deformation quantization. Lett. Math. Phys. 48(1), 35–72 (1999)

    Article  MathSciNet  MATH  Google Scholar 

  10. Lambrechts, P., Volić, I.: Formality of the little \(N\)-disks operad. Mem. Amer. Math. Soc. 230(1079), viii+116 (2014)

    MathSciNet  MATH  Google Scholar 

  11. Loday, J.L., Vallette, B.: Algebraic Operads. Grundlehren der mathematischen Wissenschaften. Springer, Berlin (2012)

    MATH  Google Scholar 

  12. Tamarkin, D.E.: Another proof of M. Kontsevich formality theorem, ArXiv Mathematics e-prints (1998)

  13. Ševera, P., Willwacher, T.: Equivalence of formalities of the little discs operad. Duke Math. J. 160(1), 175–206 (2011)

    Article  MathSciNet  MATH  Google Scholar 

  14. Vallette, B.: Algebra + homotopy = operad. In: Symplectic, Poisson, and noncommutative geometry, volume 62 of Math. Sci. Res. Inst. Publ. pp. 229–290. Cambridge University Press, New York (2014)

  15. Weibel, C.A.: An Introduction to Homological Algebra. Cambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge (1995)

    Google Scholar 

  16. Willwacher, T.: M. Kontsevich’s graph complex and the Grothendieck-Teichmüller Lie algebra. Invent. Math. 200(3), 671–760 (2014)

    Article  MathSciNet  MATH  ADS  Google Scholar 

Download references

Acknowledgements

The author is highly indebted and very thankful to his advisor Thomas Willwacher for patiently explaining most of the material presented in this text to him. He is grateful to Anton Alekseev for his constant support and numerous useful discussions. He would also like to thank Ricardo Campos, Florian Naef and Elise Raphael for many fruitful exchanges. The author was supported by the grant MODFLAT of the European Research Council (ERC).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Matteo Felder.

Appendix A. The spaces \({\mathfrak {t}}{\mathfrak {r}}_n,\mathfrak {sder}_n,\mathfrak {tder}_n\)

Appendix A. The spaces \({\mathfrak {t}}{\mathfrak {r}}_n,\mathfrak {sder}_n,\mathfrak {tder}_n\)

We follow [1]. Fix \(n\ge 1\). Let \(\mathfrak {lie}_n\) denote the completed free Lie algebra over \({\mathbb {K}}\) on n variables \(x_1,\ldots ,x_n\) and let \(\text {Ass}_n=U(\mathfrak {lie}_n)\) be the completed free associative algebra in n generators. The graded vector space of cyclic words in n variables \({\mathfrak {t}}{\mathfrak {r}}_n\) is defined as

$$\begin{aligned} {\mathfrak {t}}{\mathfrak {r}}_n:=\text {Ass}^+_n/\langle (ab-ba), a,b\in \text {Ass}_n\rangle \end{aligned}$$

where \(\text {Ass}_n^+\) is the augmentation ideal of \(\text {Ass}_n\). The Lie algebra \(\mathfrak {tder}_n\) of tangential derivations on \(\mathfrak {lie}_n\) is defined as follows. A derivation u on \(\mathfrak {lie}_n\) is tangential if there exist \(a_1,\ldots ,a_n\in \mathfrak {lie}_n\) such that \(u(x_i)=[x_i,a_i]\) for all \(i=1,\ldots ,n\). The action of u on the generators completely determine the derivation. For \(u=(a_1,\ldots ,a_n)\) and \(v=(b_1,\ldots ,b_n)\) elements of \(\mathfrak {tder}_n\), the Lie bracket is the tangential derivation \([u,v]=(c_1,\ldots ,c_n)\), where \(c_k=u(b_k)-v(a_k)+[a_k,b_k]\) for all \(k=1,\ldots ,n\). The Lie algebra of special derivations \(\mathfrak {sder}_n\) is

$$\begin{aligned} \mathfrak {sder}_n:=\left\{ u\in \mathfrak {tder}_n| u\left( \sum \limits _{i=1}^{n}{x_i}\right) =0\right\} . \end{aligned}$$

It is a Lie subalgebra of \(\mathfrak {tder}_n\). For every \(a\in \text {Ass}_n\), we have a unique decomposition

$$\begin{aligned} a=a_0+\sum \limits _{k=1}^{n}(\partial _k a)x_k, \end{aligned}$$

where \(a_0\in \mathbb {K}\) and \((\partial _k a)\in \text {Ass}_n\). The divergence map

$$\begin{aligned} \text {div}:\mathfrak {tder}_n&\rightarrow {\mathfrak {t}}{\mathfrak {r}}_n\\ u=(a_1,\ldots , a_n)&\mapsto \sum \limits _{k=1}^{n}tr(x_k(\partial _k a_k)) \end{aligned}$$

is a cocycle for \(\mathfrak {tder}_n\) ([1], Proposition 3.6.).

The following algorithm describes the isomorphism between \(H^0(\hat{gr}\mathsf {ICG}(n)^0,d_0)\), i.e., internally trivalent trees in \(\mathsf {ICG}(n)\) modulo IHX, and \(\mathfrak {sder}_n\). Let \(\Gamma \) be a tree representing an element of \(H^0(\hat{gr}\mathsf {ICG}(n)^0,d_0)\). Pick an edge incident to the external vertex 1, cut it and make it the “root” edge. The resulting tree is a binary tree with leafs labeled by \(1,\ldots , n\). Repeat this procedure for every edge incident to vertex 1, and take the sum of the trees obtained in this way. We want to interpret these binary trees as Lie words. The sign convention for this is as follows. The edges of the tree should be ordered such that its “root” edge comes first, then all edges of its left subtree, and then all edges of its right subtree. For each subtree, apply this convention recursively. The resulting linear combination of Lie words (these can be read off the trees by following the ordering of the edges) in the variables \(x_1,\ldots ,x_n\) corresponds to the first component \(a_1\) of a special derivation \(a=(a_1,\ldots ,a_n)\in \mathfrak {sder}_n\). The ith component \(a_i\) is obtained by applying the same procedure to the ith external vertex (Fig. 5).

Fig. 5
figure 5

An example of the isomorphism \(H^0(\hat{gr}\mathsf {ICG}(3)^0,d_0)\rightarrow \mathfrak {sder}_3\). The triple on the right corresponds to the element \(([x_2,x_3],-[x_1,x_3],[x_1,x_2])\)

Fig. 6
figure 6

This graph will be sent to \(tr(x_{m_1}x_{m_2}x_{m_3}x_{m_4}x_{m_5})-(-1)^5tr(x_{m_5}x_{m_4}x_{m_3}x_{m_2}x_{m_1})\) under the injective map \(H^1(\hat{gr}\mathsf {ICG}(n)^1,d_0)\rightarrow {\mathfrak {t}}{\mathfrak {r}}_n\)

We now give the map \(H^1(\hat{gr}\mathsf {ICG}(n)^1,d_0)\hookrightarrow {\mathfrak {t}}{\mathfrak {r}}_n\) as described in [13]. Let \(\overline{\Gamma }\in H^1(\hat{gr}\mathsf {ICG}(n)^1,d_0)\). We may assume that the representative \(\Gamma \) is such that the loop passes through all internal vertices. Order the edges as in Fig. 6. In this case, we map

$$\begin{aligned} \overline{\Gamma }\mapsto tr(x_{m_1}\cdots x_{m_k})-(-1)^k tr(x_{m_k}\cdots x_{m_1}). \end{aligned}$$

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Felder, M. Internally connected graphs and the Kashiwara-Vergne Lie algebra. Lett Math Phys 108, 1407–1441 (2018). https://doi.org/10.1007/s11005-018-1052-5

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11005-018-1052-5

Keywords

Mathematics Subject Classification

Navigation