AI Paper Atlas

Research map / Reinforcement learning / Offline RL

Uni[MASK]: Unified Inference in Sequential Decision Problems

Micah Carroll, Orr Paradise, Jessy Lin, Raluca Georgescu, Mingfei Sun, David Bignell, Stephanie Milani, Katja Hofmann, Matthew Hausknecht, Anca Dragan, Sam Devlin

NeurIPS 2022 · Reinforcement learning › Offline RL · approach: Decision Transformer sequence models · 2 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
Fine-tune a random-masking-trained model on the task (ablation)UniMASK · Maze2D BC, context 5reward: 2.19 -> 2.67Improved
Fine-tune a random-masking-trained model on the task (ablation)UniMASK · Maze2D RC, context 10reward: 2.31 -> 2.61Improved
Train with random masking instead of single task, no fine-tuning (ablation)UniMASK · Maze2D BC, context 5reward: 2.66 -> 2.19Hurt
Train on BC and RC jointly instead of single task (ablation)UniMASK · Maze2D BC, context 5reward: 2.66 -> 2.65No clear effect
Fine-tune multi-task model (ablation)UniMASK · Maze2D RC, context 10reward: 2.39 -> 2.42No clear effect
Use GPT backbone (Decision-GPT) instead of BERT-like UniMASK single-task (ablation)UniMASK · Maze2D RC, context 10reward: 2.41 -> 2.73Improved
Use GPT backbone instead of BERT-like at context 5 (ablation)UniMASK · Maze2D RC, context 5reward: 2.64 -> 2.32Hurt
Increase context length from 5 to 10 for UniMASK single-task (ablation)UniMASK · Maze2D BCreward: 2.66 -> 2.47Hurt
Increase context length from 5 to 10 for Decision-GPT (ablation)Decision-GPT · Maze2D BCreward: 2.66 -> 2.74Improved
Replace Decision Transformer with UniMASK fine-tunedUniMASK · Maze2D BC, context 5reward: 1.13 -> 2.67Improved
Replace feedforward NN with UniMASK single-taskUniMASK · Maze2D RC, context 5reward: 1.53 -> 2.64Improved
Random-masking training instead of single-task training (ablation)UniMASK · Minigrid validation lossvalidation lossMixed

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