AI engineering glossary · Reinforcement learning
What is HER (hindsight experience replay)?
Hindsight experience replay (also called HER): A technique for goal-based tasks that relabels failed runs as successes for the goal that was actually reached, and stores them in the buffer. This turns rare successes into frequent learning signals.
Hindsight experience replay: an example
A robot that pushed a puck to (3, 4) instead of the intended (5, 2) stores that run as if (3, 4) had been the goal.
Related terms
More Reinforcement learning terms
- Horizon
- Imitation learning
- Importance sampling ratio
- Intrinsic motivation
- Inverse reinforcement learning
- Markov decision process
- Maximum entropy reinforcement learning
- Model predictive control
Where you learn it
Hindsight experience replay is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.