Optimal control as a graphical model inference problem

DOI10.1007/s10994-012-5278-7zbMath1243.93133arXiv0901.0633OpenAlexW2107662876MaRDI QIDQ420939

Manfred Opper, Vicenç Gómez, Hilbert J. Kappen

Publication date: 23 May 2012

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/0901.0633

zbMATH Keywords

optimal controlKullback-Leibler divergenceapproximate inferencegraphical modelbelief propagationcluster variation methoduncontrolled dynamics

Mathematics Subject Classification ID

Optimal stochastic control (93E20)

Related Items

Solving high-dimensional Hamilton-Jacobi-Bellman PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space, Adaptive importance sampling for control and inference, Optimal speech motor control and token-to-token variability: a Bayesian modeling approach, Optimal design of priors constrained by external predictors, Design of biased random walks on a graph with application to collaborative recommendation, A Cost/Speed/Reliability Tradeoff to Erasing, Variational Inference for Stochastic Differential Equations, Nonlinear discrete time optimal control based on Fuzzy Models, The free energy principle made simpler but not too simple, An estimator for the relative entropy rate of path measures for stochastic differential equations, Reward Maximization Through Discrete Active Inference, Action selection in growing state spaces: control of network structure growth, Nonparametric inference of stochastic differential equations based on the relative entropy rate, A Reward-Maximizing Spiking Neuron as a Bounded Rational Decision Maker, Convergence of value functions for finite horizon Markov decision processes with constraints, Bayesian optimal control for a non-autonomous stochastic discrete time system, Planning and navigation as active inference, Generalised free energy and active inference, Sparse randomized shortest paths routing with Tsallis divergence regularization, Online control of simulated humanoids using particle belief propagation, A Minimum Free Energy Model of Motor Learning, Systems of Bounded Rational Agents with Information-Theoretic Constraints, Variational approach to rare event simulation using least-squares regression, Data assimilation: The Schrödinger perspective, On a probabilistic approach to synthesize control policies from example datasets, Learning effective state-feedback controllers through efficient multilevel importance samplers, Unnamed Item, A multilevel approach for stochastic nonlinear optimal control

Uses Software

• LibDAI

Cites Work

• Policy search for motor primitives in robotics
• Study of the starting pressure gradient in branching network
• Efficient computation of optimal actions
• Using Expectation-Maximization for Reinforcement Learning
• Dynamic programming and influence diagrams
• Constructing Free-Energy Approximations and Generalized Belief Propagation Algorithms
• Path integrals and symmetry breaking for optimal control theory
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item