Potential-Based Reward Shaping

AI and Machine Learning · Reinforcement Learning · 1999 · potential-based-reward-shaping.yaml

Add the difference of a potential function to the reward, and the optimal policy provably does not move. The result that turned reward engineering from guesswork into a statement about which extra signal is safe to give.

• supersedescorrects · extends
• classifiesspecializes · part-of
• substitutes forapproximates · alternative-to
• depends onrequires · validates

Colour is the family; a dashed line is the second member of it.

This node

• extendsadds capability to Markov Decision Processdense guidance can be added to a sparse reward without moving its optimum

References

• Policy Invariance Under Reward Transformations: Theory and Application to Reward Shaping — Ng, Harada, Russell — 1999 · link