REM

2000년 도입 · 논문 48편에서 사용

Random Ensemble Mixture (REM) is an easy to implement extension of DQN inspired by Dropout. The key intuition behind REM is that if one has access to multiple estimates of Q-values, then a weighted combination of the Q-value estimates is also an estimate for Q-values. Accordingly, in each training step, REM randomly combines multiple Q-value estimates and uses this random combination for robust training.

출처: An Optimistic Perspective on Offline Reinforcement Learning

소개 논문: An Optimistic Perspective on Offline Reinforcement Learning