Abstract
This paper considers continuous-time coordination algorithms for networks of agents that seek to collectively solve a general class of nonsmooth convex optimization problems with an inherent distributed structure. Our algorithm design builds on the characterization of the solutions of the nonsmooth convex program as saddle points of an augmented Lagrangian. We show that the associated saddle-point dynamics are asymptotically correct but, in general, not distributed because of the presence of a global penalty parameter. This motivates the design of a discontinuous saddle-point-like algorithm that enjoys the same convergence properties and is fully amenable to distributed implementation. Our convergence proofs rely on the identification of a novel global Lyapunov function for saddle-point dynamics. This novelty also allows us to identify mild convexity and regularity conditions on the objective function that guarantee the exponential convergence rate of the proposed algorithms for convex optimization problems subject to equality constraints. Various examples illustrate our discussion.
Similar content being viewed by others
References
Arrow, K., Hurwitz, L., Uzawa, H.: Studies in Linear and Non-linear Programming. Stanford University Press, Stanford (1958)
Aubin, J.P., Cellina, A.: Differential Inclusions, volume 264 of Grundlehren der Mathematischen Wissenschaften. Springer, New York (1984)
Bacciotti, A., Ceragioli, F.: Stability and stabilization of discontinuous systems and nonsmooth Lyapunov functions. ESAIM Control Optim. Calc. Var. 4, 361–376 (1999)
Bacciotti, A., Rosier, L.: Liapunov Functions and Stability in Control Theory. Communications and Control Engineering, 2nd edn. Springer, New York (2005)
Bertsekas, D.P.: Necessary and sufficient conditions for a penalty method to be exact. Math. Program. 9(1), 87–99 (1975)
Bertsekas, D.P.: Nonlinear Programming, 2nd edn. Athena Scientific, Belmont (1999)
Bertsekas, D.P., Tsitsiklis, J.N.: Parallel and Distributed Computation: Numerical Methods. Athena Scientific, Belmont (1997)
Bhat, S.P., Bernstein, D.S.: Nontangency-based Lyapunov tests for convergence and stability in systems having a continuum of equilibria. SIAM J. Control Optim. 42(5), 1745–1775 (2003)
Brogliato, B., Daniilidis, A., Lemaréchal, C., Acary, V.: On the equivalence between complementarity systems, projected systems, and differential inclusions. Syst. Control Lett. 55(1), 45–51 (2006)
Bullo, F., Cortés, J., Martínez, S.: Distributed Control of Robotic Networks. Applied Mathematics Series. Princeton University Press (2009). Electronically available at http://coordinationbook.info
Carli, R., Notarstefano, G.: Distributed partition-based optimization via dual decomposition. In: IEEE Conference on Decision and Control, pp. 2979–2984. Firenze, Italy (2013)
Chen, J., Lau, V.K.N.: Convergence analysis of saddle point problems in time varying wireless systems—control theoretical approach. IEEE Trans. Signal Process. 60(1), 443–452 (2012)
Cherukuri, A., Gharesifard, B., Cortés, J.: Saddle-point dynamics: conditions for asymptotic stability of saddle points. SIAM J. Control Optim. 55(1), 486–511 (2017)
Chiang, M., Low, S.H., Calderbank, A.R., Doyle, J.C.: Layering as optimization decomposition: a mathematical theory of network architectures. Proc. IEEE 95(1), 255–312 (2007)
Clarke, F.H.: Optimization and Nonsmooth Analysis. Canadian Mathematical Society Series of Monographs and Advanced Texts. Wiley, Hoboken (1983)
Cortés, J.: Discontinuous dynamical systems—a tutorial on solutions, nonsmooth analysis, and stability. IEEE Control Syst. 28(3), 36–73 (2008)
Dhingra, N.K., Khong, S.Z., Jovanović, M.R.: The proximal augmented Lagrangian method for nonsmooth composite optimization. IEEE Trans. Autom. Control (2018). arXiv:1610.04514
Duchi, J.C., Agarwal, A., Wainwright, M.J.: Dual averaging for distributed optimization: convergence analysis and network scaling. IEEE Trans. Autom. Control 57(3), 592–606 (2012)
Ercsey-Ravasz, M., Toroczkai, Z.: The chaos within Sudoku. Sci. Rep. 2, 725 (2012)
Feijer, D., Paganini, F.: Stability of primal-dual gradient dynamics and applications to network optimization. Automatica 46, 1974–1981 (2010)
Forti, M., Nistri, P., Quincampoix, M.: Generalized neural network for nonsmooth nonlinear programming problems. IEEE Trans. Circuits Syst. I Fundam. Theory Appl. 51(9), 1741–1754 (2004)
Gharesifard, B., Cortés, J.: Distributed convergence to Nash equilibria in two-network zero-sum games. Automatica 49(6), 1683–1692 (2013)
Gharesifard, B., Cortés, J.: Distributed continuous-time convex optimization on weight-balanced digraphs. IEEE Trans. Autom. Control 59(3), 781–786 (2014)
Goebel, R.: Stability and robustness for saddle-point dynamics through monotone mappings. Syst. Control Lett. 108, 16–22 (2017)
Henry, C.: An existence theorem for a class of differential equations with multivalued right-hand side. J. Math. Anal. Appl. 41, 179–186 (1973)
Hiriart-Urruty, J.-B., Lemaréchal, C.: Convex Analysis and Minimization Algorithms I. Grundlehren Text Editions. Springer, New York (1993)
Hiriart-Urruty, J.-B., Strodiot, J.-J., Nguyen, V.H.: Generalized Hessian matrix and second-order optimality conditions for problems with \(\cal{C}^{1,1}\) data. Appl. Math. Optim. 11, 43–56 (1984)
Holding, T., Lestas, I.: On the convergence of saddle points of concave-convex functions, the gradient method and emergence of oscillations. In: IEEE Conference on Decision and Control, pp. 1143–1148. Los Angeles, CA (2014)
Horn, R.A., Johnson, C.R.: Matrix Analysis. Cambridge University Press, Cambridge (1985)
Johansson, B., Rabi, M., Johansson, M.: A randomized incremental subgradient method for distributed optimization in networked systems. SIAM J. Control Optim. 20(3), 1157–1170 (2009)
Kelly, F.P., Maulloo, A.K., Tan, D.K.H.: Rate control in communication networks: shadow prices, proportional fairness and stability. J. Oper. Res. Soc. 49(3), 237–252 (1998)
Kose, T.: Solutions of saddle value problems by differential equations. Econometrica 24(1), 59–70 (1956)
Li, N., Zhao, C., Chen, L.: Connecting automatic generation control and economic dispatch from an optimization view. IEEE Trans. Control Netw. Syst. 3(3), 254–264 (2016)
Lu, J., Tang, C.Y.: Zero-gradient-sum algorithms for distributed convex optimization: the continuous-time case. IEEE Trans. Autom. Control 57(9), 2348–2354 (2012)
Mallada, E., Zhao, C., Low, S.H.: Optimal load-side control for frequency regulation in smart grids. IEEE Trans. Autom. Control 62(12), 6294–6309 (2017)
Mangasarian, O.L.: Sufficiency of exact penalty minimization. SIAM J. Control Optim. 23(1), 30–37 (1985)
Mateos-Núñez, D., Cortés, J.: Noise-to-state exponentially stable distributed convex optimization on weight-balanced digraphs. SIAM J. Control Optim. 54(1), 266–290 (2016)
Nagurney, A., Zhang, D.: Projected Dynamical Systems and Variational Inequalities with Applications. International Series in Operations Research and Management Science, vol. 2. Kluwer Academic Publishers, Dordrecht (1996)
Necoara, I., Nedelcu, V.: Rate analysis of inexact dual first order methods. IEEE Trans. Autom. Control 59(5), 1232–1243 (2014)
Necoara, I., Suykens, J.: Application of a smoothing technique to decomposition in convex optimization. IEEE Trans. Autom. Control 53(11), 2674–2679 (2008)
Nedic, A., Ozdaglar, A.: Distributed subgradient methods for multi-agent optimization. IEEE Trans. Autom. Control 54(1), 48–61 (2009)
Nedic, A., Ozdaglar, A., Parrilo, P.A.: Constrained consensus and optimization in multi-agent networks. IEEE Trans. Autom. Control 55(4), 922–938 (2010)
Niederländer, S.K., Cortés, J.: Distributed coordination for separable convex optimization with coupling constraints. In: IEEE Conference on Decision and Control, pp. 694–699. Osaka, Japan (2015)
Niederländer, S.K., Allgöwer, F., Cortés, J.: Exponentially fast distributed coordination for nonsmooth convex optimization. In: IEEE Conference on Decision and Control, pp. 1036–1041. Las Vegas, NV (2016)
Pappalardo, M., Passacantando, M.: Stability for equilibrium problems: from variational inequalities to dynamical systems. J. Optim. Theory Appl. 113(3), 567–582 (2002)
Polyak, B.T.: Iterative methods using Lagrange multipliers for solving extremal problems with constraints of the equation type. USSR Comput. Math. Math. Phys. 10(5), 1098–1106 (1970)
Ratliff, L.J., Burden, S.A., Sastry, S.S.: On the characterization of local Nash equilibria in continuous games. IEEE Trans. Autom. Control 61(8), 2301–2307 (2016)
Richert, D., Cortés, J.: Robust distributed linear programming. IEEE Trans. Autom. Control 60(10), 2567–2582 (2015)
Rockafellar, R.T.: Convex Analysis. Princeton University Press, Princeton (1970)
Rockafellar, R.T., Wets, R.J.B.: Variational Analysis, Volume of 317 Comprehensive Studies in Mathematics. Springer, New York (1998)
Shi, W., Ling, Q., Wu, G., Yin, W.: EXTRA: an exact first-order algorithm for decentralized consensus optimization. SIAM J. Optim. 25(2), 944–966 (2015)
Stegink, T., Persis, C.D., van der Schaft, A.J.: A unifying energy-based approach to stability of power grids with market dynamics. IEEE Trans. Autom. Control 62(6), 2612–2622 (2017)
Wan, P., Lemmon, M.D.: Event-triggered distributed optimization in sensor networks. In: Symposium on Information Processing of Sensor Networks, pp. 49–60., San Francisco, CA (2009)
Wang, J., Elia, N.: A control perspective for centralized and distributed convex optimization. In: IEEE Conference on Decision and Control, pp. 3800–3805. Orlando, Florida (2011)
Zhang, X., Papachristodoulou, A.: A real-time control framework for smart power networks: design methodology and stability. Automatica 58, 43–50 (2015)
Zhao, C., Topcu, U., Li, N., Low, S.H.: Design and stability of load-side primary frequency control in power systems. IEEE Trans. Autom. Control 59(5), 1177–1189 (2014)
Zhu, M., Martínez, S.: On distributed convex optimization under inequality and equality constraints. IEEE Trans. Autom. Control 57(1), 151–164 (2012)
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Anthony Bloch.
This work was supported by the German National Academic Foundation, the Dr. Jürgen and Irmgard Ulderup Foundation, Award FA9550-15-1-0108, and NSF Award CNS-1329619. Preliminary versions of this manuscript appeared as Niederländer et al. (2016), Niederländer and Cortés (2015) at the IEEE Conference on Decision and Control.
Appendix
Appendix
We gather in this appendix various intermediate results used in the derivation of the main results of the paper.
1.1 Auxiliary Results for Performance Characterization
The following two results characterize properties of the generalized Hessian of \(\mathcal {C}^{1,1}\) functions. In each case, let \(f\in \mathcal {C}^{1,1}({\mathbb R}^{n},\mathbb R)\) and consider the set-valued map \(\partial (\nabla f):{\mathbb R}^{n}\rightrightarrows {\mathbb R}^{n \times n}\) as defined in Sect. 2.1. For \(x,y\in {\mathbb R}^{n}\), we let \([x,y]=\{x+\theta (y-x)\mid \theta \in [0,1]\}\) and study the set
Lemma A.1
(Positive definiteness) Let \(f\in \mathcal {C}^{1,1}({\mathbb R}^{n},\mathbb R)\) and suppose \(\partial (\nabla f)\succ 0\). Then, \({{\mathrm{co}}}\big \{\partial (\nabla f([x,y]))\big \}\succ 0\) for all \(x,y\in {\mathbb R}^{n}\).
Proof
Since \(\partial (\nabla f)(x)\succ 0\) for all \(x\in {\mathbb R}^{n}\), it follows \(\bigcup _{z\in [x,y]}\partial (\nabla f)(z)\succ 0\). Moreover, since the cone of symmetric positive definite matrices is convex itself, it contains all convex combinations of elements in \(\bigcup _{z\in [x,y]}\partial (\nabla f)(z)\), i.e., in particular \({{\mathrm{co}}}\big \{\partial (\nabla f([x,y]))\big \}\succ 0\), concluding the proof.\(\square \)
Lemma A.2
(Compactness) Let \(f\in \mathcal {C}^{1,1}({\mathbb R}^{n},\mathbb R)\). Then, the set \(\partial (\nabla f([x,y]))\) and its convex closure are both compact.
Proof
Boundedness of the set \(\partial (\nabla f([x,y]))\) follows from Rockafellar and Wets (1998, Proposition 5.15). To show that \(\partial (\nabla f([x,y]))\) is closed, take \(\nu \in {{\mathrm{cl}}}\partial (\nabla f([x,y]))\). By definition, there exists \(\{\nu _{n}\}\subset \partial (\nabla f([x,y]))\) such that \(\nu _{n}\rightarrow \nu \). Since \(\nu _{n}\) belongs to \(\partial (\nabla f([x,y]))\), let us denote \(\nu _{n}\in \partial (\nabla f)(z_{n})\), i.e., \(\nu _{n}\) is a vector based at \(z_{n}\). Similarly, let z be the point at which the vector \(\nu \) is based. Following the above arguments, we have \(z_{n}\rightarrow z\). Since [x, y] is compact and \(z_{n}\in [x,y]\), we deduce \(z\in [x,y]\). Assume, by contradiction, that \(\nu \notin \partial (\nabla f)(z)\). Then, since \(\partial (\nabla f)(z)\) is closed, there exists \(\varepsilon >0\) such that \(\{\nu \}\cap \partial (\nabla f)(z)+\mathbb {B}(0,\varepsilon )=\emptyset \). Using upper semi-continuity, there exists \(N\in \mathbb {N}\) such that if \(n\ge N\), then \(\partial (\nabla f)(z_{n})\subset \partial (\nabla f)(z)+\mathbb {B}(0,\varepsilon )\). This fact is in contradiction with \(\nu _{n}\rightarrow \nu \). Therefore, it follows \(\nu \in \partial (\nabla f)(z)\subset \partial (\nabla f([x,y]))\), and we conclude \(\partial (\nabla f([x,y]))\) is closed. Thus, \(\partial (\nabla f([x,y]))\) is compact, and so is \({{\mathrm{co}}}\big \{\partial (\nabla f([x,y]))\big \}\), concluding the proof.\(\square \)
1.2 Auxiliary Results for Convergence Analysis of Saddle-Point-Like Dynamics
Here, we investigate the explicit computation of the projection operator \(P_{T_G}\), which plays a key role in the dynamics (SPLD). Recall that \(T_{G}(x)\) and \(N_{G}(x)\) denote the tangent and normal cone of \(G\subset {\mathbb R}^{n}\) at \(x\in G\), respectively. The following geometric interpretation of \(P_{T_G}\) is well known in the literature of locally projected dynamical systems (Nagurney and Zhang 1996; Pappalardo and Passacantando 2002):
-
(i)
if \((x,\lambda )\in {{\mathrm{int}}}G\times {\mathbb R}^{p}\), then \(P_{T_G(x)}(F(x,\lambda ))=F(x,\lambda )\);
-
(ii)
if \((x,\lambda )\in {{\mathrm{bd}}}G\times {\mathbb R}^{p}\), then
$$\begin{aligned} P_{T_G(x)}(F(x,\lambda ))=\bigcup _{\xi \in F(x,\lambda )}\xi -\max \big \{0, \langle \xi ,n^{\star }(x,\xi )\rangle \big \}n^{\star }(x,\xi ), \end{aligned}$$where
$$\begin{aligned} n^{\star }(x,\xi )\in \mathop \mathrm{argmax}_{n\in N_{G}^{\sharp }(x)}\langle \xi ,n\rangle . \end{aligned}$$(12)
Note that if \(\{\xi \}\cap T_{G}(x)\ne \emptyset \) for some \((x,\lambda )\in {{\mathrm{bd}}}G\times {\mathbb R}^{p}\) and \(\xi \in F(x,\lambda )\), then \(\sup _{n\in N_{G}^{\sharp }(x)}\langle \xi ,n\rangle \le 0\), and by definition of \(P_{T_G}\), no projection needs to be performed. The following result establishes the existence and uniqueness of the maximizer \(n^{\star }(x,\xi )\) of (12) whenever \(\{\xi \}\cap T_{G}(x)=\emptyset \).
Lemma A.3
(Computation of the projection operator) Let \((x,\lambda )\in {{\mathrm{bd}}}G\times {\mathbb R}^{p}\). If there exists \(\xi \in F(x,\lambda )\) such that \(\sup _{n\in N_{G}^{\sharp }(x)}\langle \xi ,n\rangle >0\), then the maximizer \(n^{\star }(x,\xi )\) of (12) exists and is unique.
Proof
Let \((x,\lambda )\in {{\mathrm{bd}}}G\times {\mathbb R}^{p}\) and suppose there exists \(\xi \in F(x,\lambda )\) such that \(\sup _{n\in N_{G}^{\sharp }(x)}\langle \xi ,n\rangle >0\). By definition, the normal cone \(N_{G}(x)\) of G at \(x\in {{\mathrm{bd}}}G\) is closed and convex. Existence of \(n^{\star }(x,\xi )\) follows from compactness of the set \(N_{G}^{\sharp }(x)\). Now, let \(\tilde{n}^{\star }(x,\xi )\) and \(\hat{n}^{\star }(x,\xi )\) be two distinct maximizers of (12) such that \(\langle \xi ,\tilde{n}^{\star }(x,\xi )\rangle >0\) and \(\langle \xi ,\hat{n}^{\star }(x,\xi )\rangle >0\). Convexity implies \((\tilde{n}^{\star }(x,\xi ) + \hat{n}^{\star }(x,\xi ))/||\tilde{n}^{\star }(x,\xi )+\hat{n}^{\star }(x,\xi )||\in N_{G}^{\sharp }(x)\). Therefore, it follows
which contradicts the fact that \(\tilde{n}^{\star }(x,\xi )\) maximizes (12).\(\square \)
We note that the computational complexity of solving (12) depends not only on the problem dimensions \(n,p,m>0\), but also on the convexity and regularity assumptions of the problem data, i.e., on f, h and g.
Rights and permissions
About this article
Cite this article
Cortés, J., Niederländer, S.K. Distributed Coordination for Nonsmooth Convex Optimization via Saddle-Point Dynamics. J Nonlinear Sci 29, 1247–1272 (2019). https://doi.org/10.1007/s00332-018-9516-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00332-018-9516-4
Keywords
- Distributed multi-agent coordination
- Nonsmooth convex optimization
- Saddle-point dynamics
- Continuous-time optimization algorithms