AI Paper Atlas

Research map / Reinforcement learning / Offline RL

SPQR: Controlling Q-ensemble Independence with Spiked Random Model for Reinforcement Learning

Dohyeok Lee, Seungyub Han, Taehyun Cho, Jungwoo Lee

NeurIPS 2023 · Reinforcement learning › Offline RL · approach: Pessimism and uncertainty in offline RL · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add SPQR independence regularization to SAC-MinSAC-Min · D4RL Gym averagenormalized average return: 84.5 -> 89.6Improved
Add SPQR independence regularization to CQL-MinCQL-Min · Franka Kitchen averagenormalized average return: 48.2 -> 52.6Improved
Add SPQR independence regularization to CQL-MinCQL-Min · Antmaze averagenormalized average return: 73.1 -> 84.4Improved
Add SPQR regularization with β=0.1 to SAC-MinSAC-Min · hopper-random (evaluated on hopper-full-replay)performance: 31.3±0.0 -> 35.6±1.4Improved
Increase SPQR regularization gain from β=0.1 to β=100.0 (ablation)SPQR-SAC-Min · hopper-random (evaluated on hopper-full-replay)spike count: 2182 -> 698Improved
Increase SPQR regularization gain from β=0.1 to β=100.0 (ablation)SPQR-SAC-Min · hopper-random (evaluated on hopper-full-replay)independence test acceptance ratio: 80.4% -> 97.6%Improved
Increase SPQR regularization gain from β=0.1 to β=100.0 (ablation)SPQR-SAC-Min · hopper-random (evaluated on hopper-full-replay)performance: 35.6±1.4 -> 32.9±0.8Hurt
Use β=0.01 instead of β=0.1 (ablation)SPQR-SAC-Min · hopper-random (evaluated on hopper-full-replay)performance: 35.6±1.4 -> 33.6±1.8Hurt
Use β=1.0 instead of β=0.1 (ablation)SPQR-SAC-Min · hopper-random (evaluated on hopper-full-replay)performance: 35.6±1.4 -> 32.6±0.4Hurt
Use β=10.0 instead of β=0.1 (ablation)SPQR-SAC-Min · hopper-random (evaluated on hopper-full-replay)performance: 35.6±1.4 -> 32.7±0.4Hurt
Add SPQR independence regularization to EDACEDAC · D4RL Gym tasksnormalized average returnMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy