IRCoCo: Immediate Rewards-Guided Deep Reinforcement Learning for Code Completion.
Bolun LiZhihong SunTao HuangHongyu ZhangYao WanGe LiZhi JinChen LyuPublished in: Proc. ACM Softw. Eng. (2024)
Keyphrases
- reinforcement learning
- function approximation
- markov decision processes
- state space
- reinforcement learning algorithms
- learning algorithm
- model free
- reward function
- reward shaping
- source code
- multi agent
- neural network
- temporal difference learning
- deep learning
- dynamic programming
- action selection
- learning process
- static analysis
- optimal control
- transfer learning
- optimal policy
- complex domains
- markov decision process
- mobile robot
- control policy
- policy search
- bandit problems