Approximate Linear Programming for Solving Hybrid Factored MDPs
Explore this paper's citation graph
Summary
This work presents the HALP framework and discusses several representational and computational issues that make the approach appropriate for large MDPs, and demonstrates the feasibility of the approach on high-dimensional distributed control problems.
- Type
- article
- Published
- 2006-01-01
- Cited by
- 16
- References
- 11
- OpenAlex
- https://openalex.org/W75459635
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:16326272
Keywords
Markov decision process, Mathematical optimization, Bellman equation, Linear programming, Computer science
References
- An MCMC Approach to Solving Hybrid Factored MDPs
- The Linear Programming Approach to Approximate Dynamic Programming
- A model for reasoning about persistence and causation
- Generalized polynomial approximations in Markovian decision processes
- Max-norm Projections for Factored MDPs
- Linear Program Approximations for Factored Continuous-State Markov Decision Processes
- Direct value-approximation for factored MDPs
- On Constraint Sampling in the Linear Programming Approach to Approximate Dynamic Programming
- Solving Factored MDPs with Continuous and Discrete Variables
- Efficient Solution Algorithms for Factored MDPs
- Methods of Mathematical Physics
- Efficient Solution Algorithms for Factored MDPs
Cited by
- Multi-Objective Constraint Satisfaction for Mobile Robot Area Defense
- Partially Observable Markov Decision Processes (POMDPs) and Wireless Body Area Networks (WBAN): A Survey
- Integrated resource allocation and planning in stochastic multiagent environments
- On the locality of action domination in sequential decision making
- Extending the Bellman equation for MDPs to continuous actions and cont. time in the discounted case
- Representation and generalization in autonomous reinforcement learning
- Problèmes décisionnels de Markov temporels : formalisation et résolution
- Approximate Policy Iteration for Generalized Semi-Markov Decision Processes: an Improved Algorithm
- XMDP : un modèle de planification temporelle dans l'incertain à actions paramétriques
- Temporal Markov Decision Problems : Formalization and Resolution
- Un Algorithme Amélioré d'Itération de la Politique Approchée pour les Processus Décisionnels Semi-Markoviens Généralisés
- Temporal coordination under uncertainty: initial resultsfor the two agents case
- Constraint-Generation Policy Optimization (CGPO): Nonlinear Programming for Policy Optimization in Mixed Discrete-Continuous MDPs
- Overview of Linear Program Approximations for Factored Continuous and Hybrid-State Markov Decision Processes
- A Simulation-based Approach for Solving Generalized Semi-Markov Decision Processes
- Temporal coordination under uncertainty: initial results for the two agents case
Related papers
- Learning Basis Functions in Hybrid Domains
- On constraint sampling in the linear programming approach to approximate linear programming
- Revisiting Approximate Linear Programming Using a Saddle Point Based Reformulation and Root Finding Solution Approach
- Revisiting Approximate Linear Programming: Constraint-Violation Learning with Applications to Inventory Control and Energy Storage
- Randomized Linear Programming Solves the Markov Decision Problem in Nearly Linear (Sometimes Sublinear) Time
- Network-Based Approximate Linear Programming for Discrete Optimization
- Towards Exploiting Duality in Approximate Linear Programming for MDPs