Large Language Models can Implement Policy Iteration.
Ethan A. BrooksLogan WallsRichard L. LewisSatinder SinghPublished in: NeurIPS (2023)
Keyphrases
- language model
- policy iteration
- markov decision processes
- language modeling
- model free
- reinforcement learning
- fixed point
- optimal policy
- n gram
- sample path
- probabilistic model
- query expansion
- language modelling
- document retrieval
- speech recognition
- least squares
- information retrieval
- retrieval model
- markov decision process
- statistical language models
- finite state
- temporal difference
- test collection
- state space
- language models for information retrieval
- smoothing methods
- infinite horizon
- relevance model
- optimal control
- machine learning
- convergence rate
- data mining
- linear programming
- spoken term detection