architecture · first seen 20 Oct 2026
Decision-Pretrained Transformer
Also: DPT
A transformer pretrained on decision data to perform in-context reinforcement learning across bandit and MDP tasks.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Is In-Context Reinforcement Learning A Posterior Sampler? An Empirical StudyPapers · 20 Oct 2026 · Weiqin Chen, Xinlong Du, Gesualdo Scutari, Harsha Honnappa
Latest sources
- Is In-Context Reinforcement Learning A Posterior Sampler? An Empirical StudyPapers · 20 Oct 2026 · Weiqin Chen, Xinlong Du, Gesualdo Scutari, Harsha Honnappa