B-Coder: Value-Based Deep Reinforcement Learning for Program Synthesis.
Zishun YuYunzhe TaoLiyu ChenTao SunHongxia YangPublished in: ICLR (2024)
Keyphrases
- program synthesis
- reinforcement learning
- function approximation
- subband
- image compression
- learning algorithm
- compression ratio
- model free
- state space
- recursive programs
- markov decision processes
- inductive logic programming
- multiresolution
- image coder
- bitstream
- image coding
- supervised learning
- high level
- search engine