Open AccessPosted Content
Mean-field Markov decision processes with common noise and open-loop controls
Médéric Motte,Huyên Pham +1 more
Reads0
Chats0
TLDR
The correspondence between CMKV-MDP and a general lifted MDP on the space of probability measures is proved, and the dynamic programming Bellman fixed point equation satisfied by the value function is established.Abstract:
We develop an exhaustive study of Markov decision process (MDP) under mean field interaction both on states and actions in the presence of common noise, and when optimization is performed over open-loop controls on infinite horizon. Such model, called CMKV-MDP for conditional McKean-Vlasov MDP, arises and is obtained here rigorously with a rate of convergence as the asymptotic problem of N-cooperative agents controlled by a social planner/influencer that observes the environment noises but not necessarily the individual states of the agents. We highlight the crucial role of relaxed controls and randomization hypothesis for this class of models with respect to classical MDP theory. We prove the correspondence between CMKV-MDP and a general lifted MDP on the space of probability measures, and establish the dynamic programming Bellman fixed point equation satisfied by the value function, as well as the existence of-optimal randomized feedback controls. The arguments of proof involve an original measurable optimal coupling for the Wasserstein distance. This provides a procedure for learning strategies in a large population of interacting collaborative agents. MSC Classification: 90C40, 49L20.read more
Citations
More filters
Journal ArticleDOI
Old and New
TL;DR: Richard Mateosian reviews old and new books, including Weinberg on Writing--The Fieldstone Method, The Art of Computer Programming, From Java to Ruby--Things Every Manager Should Know, and Introduction to DITA--A User Guide to the Darwin Information Typing Architecture.
Posted Content
An Overview of Multi-Agent Reinforcement Learning from Game Theoretical Perspective
Yaodong Yang,Jun Wang +1 more
TL;DR: This work provides a self-contained assessment of the current state-of-the-art MARL techniques from a game theoretical perspective and expects this work to serve as a stepping stone for both new researchers who are about to enter this fast-growing domain and existing domain experts who want to obtain a panoramic view and identify new directions based on recent advances.
Posted Content
Model-Free Mean-Field Reinforcement Learning: Mean-Field MDP and Mean-Field Q-Learning
TL;DR: This work introduces generic model-free algorithms based on the state-action value function at the mean field level and proves convergence for a prototypical Q-learning method for mean field control problems.
Numerical methods for mean field games and mean field type control
TL;DR: Numerical schemes for forward-backward systems of partial differential equations (PDEs), optimization techniques for variational problems driven by a Kolmogorov-Fokker-Planck PDE, an approach based on a monotone operator viewpoint, and stochastic methods relying on machine learning tools are discussed.
References
More filters
Book
Reinforcement Learning: An Introduction
TL;DR: This book provides a clear and simple account of the key ideas and algorithms of reinforcement learning, which ranges from the history of the field's intellectual foundations to the most recent developments and applications.
Book
Dynamic Programming and Optimal Control
TL;DR: The leading and most up-to-date textbook on the far-ranging algorithmic methododogy of Dynamic Programming, which can be used for optimal control, Markovian decision problems, planning and sequential decision making under uncertainty, and discrete/combinatorial optimization.
Book
Foundations of modern probability
TL;DR: In this article, the authors discuss the relationship between Markov Processes and Ergodic properties of Markov processes and their relation with PDEs and potential theory. But their main focus is on the convergence of random processes, measures, and sets.
BookDOI
Mass transportation problems
TL;DR: In this article, a modification of the Monge-Kantorovich Problem with relaxed or additional constraints is presented. But this modification is restricted to the case where the Kantorovich-type metrics are applied to various Probabilistic-Type Limit Theorems.
Book
Mean Field Games and Mean Field Type Control Theory
TL;DR: In this article, the authors present a general presentation of mean field control problems in the mean field game and the Mean Field Type Control problem in Nash games with a large number of players.