AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Is Model Ensemble Necessary? Model-based RL via a Single Model with Lipschitz Regularized Value Function

Ruijie Zheng, Xiyao Wang, Huazhe Xu, Furong Huang

ICLR 2023 · Reinforcement learning › Policy gradient and actor-critic · approach: Model-based RL with guarantees · 0 citations (OpenAlex)

Read the paper

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 robust regularization to the value network while using a single deterministic dynamics modelMBPO · Walker, HalfCheetah, Ant, Humanoid, HopperreturnImproved
Use robust regularization instead of a probabilistic model ensembleMBPO · Walkerwall-clock time: 60.0 -> 45.4 hImproved
Use robust regularization instead of a probabilistic model ensembleMBPO · Antwall-clock time: 62.3 -> 48.3 hImproved
Add random uniform noise instead of adversarially selecting perturbations (ablation)Robust Regularization on MBPO · WalkerreturnHurt
Use PGD-20 instead of FGSM to compute perturbations (ablation)Robust Regularization on MBPO · WalkerreturnNo clear effect
Vary the robust-regularization weight λ (ablation)Robust Regularization on MBPO · WalkerreturnMixed
Use a tighter global spectral-normalization constraint (ablation)Spectral Normalization on MBPO · WalkerreturnMixed
Combine spectral normalization with a single probabilistic model rather than a single deterministic model (ablation)MBPO · AntreturnImproved
Combine robust regularization with a single probabilistic model rather than a single deterministic model (ablation)MBPO · AntreturnImproved
Add Lipschitz regularization to model-free SAC (ablation)SAC · MuJoCo tasksreturnImproved

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