Adaptive Reward Computation in Reinforcement Learning-Based Continuous Integration Testing.
Yang YangChaoyue PanZheng LiRuilian ZhaoPublished in: IEEE Access (2021)
Keyphrases
- reinforcement learning
- integration testing
- function approximation
- action space
- black box
- eligibility traces
- object oriented programs
- partially observable environments
- model free
- reward function
- optimal policy
- partially observable markov decision processes
- temporal difference
- average reward
- markov decision processes
- reinforcement learning methods
- learning process
- fitted q iteration
- state action
- action selection
- function approximators
- inverse reinforcement learning
- component based software
- learning capabilities
- reinforcement learning algorithms
- software testing
- test cases
- dynamic programming
- database
- control policy
- state space
- multi agent
- case study
- information systems
- learning algorithm
- machine learning
- real world