AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Playing hard exploration games by watching YouTube

Yusuf Aytar, Tobias Pfaff, David Budden, Thomas Paine, Ziyu Wang, Nando de Freitas

NeurIPS 2018 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Intrinsically motivated exploration · 6 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
Guide IMPALA with TDC+CMC checkpoints from a YouTube video, without environment rewardIMPALA · Montezuma’s Revengescore: 37,232.7Improved
Guide IMPALA with TDC+CMC checkpoints from a YouTube video, without environment rewardIMPALA · Pitfall!score: 54,912.4Improved
Guide IMPALA with TDC+CMC checkpoints from a YouTube video, without environment rewardIMPALA · Private Eyescore: 98,212.5Improved
Add environment reward alongside the TDC+CMC imitation reward (ablation)YouTube imitation IMPALA · Montezuma’s Revengescore: 37,232.7 -> 58,175.1Improved
Add environment reward alongside the TDC+CMC imitation reward (ablation)YouTube imitation IMPALA · Pitfall!score: 54,912.4 -> 76,812.5Improved
Add environment reward alongside the TDC+CMC imitation reward (ablation)YouTube imitation IMPALA · Private Eyescore: 98,212.5 -> 98,763.2Improved
Replace pixel-space features with combined TDC+CMC embeddings for imitation reward (ablation)IMPALA imitation agent · Montezuma’s Revenge, Pitfall!, Private Eyegame scoreImproved
Use only pixel-space features for imitation reward instead of TDC+CMC embeddings (ablation)IMPALA imitation agent · Private EyescoreImproved
Use only pixel-space features for imitation reward instead of TDC+CMC embeddings (ablation)IMPALA imitation agent · Montezuma’s Revenge, Pitfall!game progressHurt
Train the policy using frozen pretrained TDC+CMC features instead of the full image-based policy network (ablation)IMPALA · Montezuma’s RevengescoreHurt
Condition the feature-based policy on the next checkpoint embedding (ablation)IMPALA trained from pretrained features · Montezuma’s Revengelearning speedImproved
Change the YouTube expert trajectory among the three tested Montezuma videos (ablation)YouTube imitation IMPALA · Montezuma’s RevengescoreMixed

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