AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Model-Based Reinforcement Learning With Isolated Imaginations

Minting Pan, Xiangming Zhu, Yitao Zheng, Yunbo Wang, Xiaokang Yang

TPAMI 2023 · Reinforcement learning › Visual and representation RL · approach: Inferring decision-making from behavior

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
use Iso-Dream++ instead of DreamerV2 and Denoised-MDPIso-Dream++ · CARLA (500k steps)average return: around 60 vs 10 (DreamerV2) and 25 (Denoised-MDP)Improved
add inverse dynamics (Inverse Cell) (ablation)Iso-Dream++ · CARLAreturnImproved
add rollouts of future noncontrollable states as action model input (ablation)Iso-Dream++ · CARLAreturnImproved
add future-state attention instead of concatenation (ablation)Iso-Dream++ · CARLAreturnImproved
add separate static-information branch (ablation)Iso-Dream++ · CARLAreturn: decreases by about 15% without itImproved
add min-max variance constraints and sparse dependency modeling (ablation)Iso-Dream++ · CARLAreturnImproved
add sparse dependency modeling (dependency gate) (ablation)Iso-Dream++ · CARLAreturnImproved
rollout window tau 5 vs 10 vs 15 (ablation)Iso-Dream++ · CARLAreturn: no remarkable differences among tau in 5, 10, 15No clear effect
add rollout of noncontrollable states (tau 5 vs tau 0) (ablation)Iso-Dream++ · CARLAreturn: tau=0 significantly worse than tau in 5, 10, 15Improved
add min-max variance constraints (ablation)Iso-Dream++ · DMC video_easy (10 seeds)return: improves most tasks, especially Finger SpinImproved
evaluate on video_hard after training on video_easy: Iso-Dream++ vs DreamerProIso-Dream++ · DMC train video_easy test video_hardreturn: Finger Spin 628 -> 692; Hopper Stand 180 -> 643; Walker Walk 533 -> 642; Cheetah Run 244 -> 441Improved
transfer pretrained action-free branch from Day to Night modeIso-Dream++ · CARLA Day to NightreturnImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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