A Deterministic Annealing Optimization Approach for Witsenhausen's and Related Decentralized Control Settings
Authors:
- Mustafa MehmetogluMMMustafa Mehmetoglu
- Emrah AkyolEAEmrah Akyol
- Kenneth RoseKRKenneth Rose
Preprint
Abstract
This paper studies the problem of mapping optimization in decentralized control problems. A global optimization algorithm is proposed based on the ideas of ``deterministic annealing" - a powerful non-convex optimization framework derived from information theoretic principles with analogies to statistical physics. The key idea is to randomize the mappings and control the Shannon entropy of the system during optimization. The entropy constraint is gradually relaxed in a deterministic annealing process while tracking the minimum, to obtain the ultimate deterministic mappings. Deterministic annealing has been successfully employed in several problems including clustering, vector quantization, regression, as well as the Witsenhausen's counterexample in our recent work[1]. We extend our method to a more involved setting, a variation of Witsenhausen's counterexample, where there is a side channel between the two controllers. The problem can be viewed as a two stage cancellation problem. We demonstrate that there exist complex strategies that can exploit the side channel efficiently, obtaining significant gains over the best affine and known non-linear strategies.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- A deterministic annealing approach to Witsenhausen's counterexample2014
- Network Information Theory2011
- Numerical solutions to the Witsenhausen counterexample by approximating networksIEEE Transactions on Automatic Control, 2001
- The Witsenhausen counterexample: a hierarchical search approach for nonconvex optimization problemsIEEE Transactions on Automatic Control, 2001
- An ordinal optimization approach to optimal control problemsAutomatica, 1999
- Deterministic annealing for clustering, compression, classification, regression, and related optimization problemsProceedings of the IEEE, 1998
- Statistical mechanics and phase transitions in clusteringPhysical Review Letters, 1990
- Source coding with side information and a converse for degraded broadcast channelsIEEE Transactions on Information Theory, 1975
- A Counterexample in Stochastic Optimum ControlSIAM Journal on Control, 1968