Near-Optimality of Finite-Memory Codes and Reinforcement Learning for Zero-Delay Coding of Markov Sources.
Liam CreggFady AlajajiSerdar YükselPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- error control
- decoding complexity
- error correcting
- decoding algorithm
- intersymbol interference
- state and action spaces
- coding scheme
- error correction
- markov chain
- amplitude modulation
- low memory
- block codes
- function approximation
- optimal solution
- logical operations
- data sources
- coding method
- main memory
- parity check
- learning algorithm
- information sources
- turbo codes
- average reward
- random access
- markov model
- markov decision processes
- memory requirements
- low density parity check
- reed solomon
- finite number
- state space
- multiple sources
- optimal policy
- multipath
- conditional independence
- model free
- dynamic programming
- inter frame
- reinforcement learning algorithms
- markov decision process