日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

週刊ニュースレター購読
強化学習/世界モデルarXiv:2510.21418

DreamerV3-XP:不確実性推定による探索の最適化

DreamerV3-XP: Optimizing exploration through uncertainty estimation

シェア:XThreadsFacebookLINEはてブBluesky

世界モデル強化学習DreamerV3に優先経験再生とアンサンブル不確実性に基づく内的報酬を追加し、疎な報酬環境での探索と学習効率を改善した。

著者: Lukas Bierling, Davide Pasero, Jan-Henrik Bertrand, Kiki Van Gerwen

分類: cs.LG, cs.AI

原文アブストラクト

We introduce DreamerV3-XP, an extension of DreamerV3 that improves exploration and learning efficiency. This includes (i) a prioritized replay buffer, scoring trajectories by return, reconstruction loss, and value error and (ii) an intrinsic reward based on disagreement over predicted environment rewards from an ensemble of world models. DreamerV3-XP is evaluated on a subset of Atari100k and DeepMind Control Visual Benchmark tasks, confirming the original DreamerV3 results and showing that our extensions lead to faster learning and lower dynamics model loss, particularly in sparse-reward settings.

関連論文