# Reproduction: Taming Aleatoric Impulse in Off-Policy Reinforcement Learning

[OpenReview paper](https://openreview.net/forum?id=2tcZ8Vl30Q) · [Official code](https://github.com/yzy-yuzhouyang/DSAC-AID)

## Pages

| Page |
| --- |
| [Executive summary](#/executive-summary) |
| [Claim 1: SOTA on Gym-MuJoCo and DMC](#/claim-1-sota-on-gym-mujoco-and-dmc) |
| [Claim 2: Epistemic vs aleatoric disentanglement](#/claim-2-epistemic-vs-aleatoric-disentanglement) |
| [Claim 3: LCB pessimism and UCB exploration](#/claim-3-lcb-pessimism-and-ucb-exploration) |
| [Conclusion](#/conclusion) |