A constant bias added to a potential-based reward shaping function, set to compensate the external reward and the initial Q-values, improves sample efficiency while preserving the optimal policy.
Barto, Richard S
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Improving the Effectiveness of Potential-Based Reward Shaping in Reinforcement Learning
A constant bias added to a potential-based reward shaping function, set to compensate the external reward and the initial Q-values, improves sample efficiency while preserving the optimal policy.