On Structural Properties of MDPs that Bound Loss Due to Shallow Planning.
Nan JiangSatinder P. SinghAmbuj TewariPublished in: IJCAI (2016)
Keyphrases
- action selection
- structural properties
- reinforcement learning
- markov decision processes
- topological properties
- lower bound
- state space
- upper bound
- tree width
- natural language processing
- worst case
- information extraction
- optimal policy
- partially observable
- network evolution
- learning algorithm
- decision theoretic planning
- stochastic domains
- markov decision problems
- machine learning
- probabilistic planning
- finite horizon
- planning problems
- question answering
- sequential decision making problems
- factored markov decision processes