CAPE: 対照的アクション条件付き並列符号化による身体化プランニング
CAPE: Contrastive Action-conditioned Parallel Encoding for Embodied Planning
ロボットの行動計画のため、アクション系列の将来結果を区別する対照学習を用いた視覚ダイナミクスモデルを提案し、実世界とシミュレーションで性能を向上させた。
詳しい要約
1. どんなもの？
2. 先行研究と比べてどこがすごい？
3. 技術・手法の肝は？
4. どうやって有効だと検証した？
5. 議論はある？
6. 次に読むべき論文は？
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Cong Chen, Haowen Wang, Zhixiang Zhang, Pei Ren, Zhengping Che
分類: cs.RO
原文アブストラクト
Embodied agents need to predict the future consequences of candidate actions in order to plan effectively before execution. Existing visual dynamics models learn by reconstructing future visual states or rolling out dense latent representations, which spreads learning capacity across visually salient but planning-irrelevant content rather than the action-conditioned changes that drive manipulation outcomes. We propose CAPE, a Contrastive Action-conditioned Parallel Encoding framework that learns visual dynamics by distinguishing the future outcomes induced by different action sequences. Given an initial observation and a candidate action sequence, CAPE decodes the full future latent trajectory in a single forward pass and is trained with a Goal-Convergent Contrastive Objective that aligns predictions corresponding to the same future outcome while separating those corresponding to different outcomes. On real-world DROID and zero-shot transfer to RoboCasa, CAPE substantially outperforms prior baselines on future-state retrieval, offline action matching, and closed-loop planning, while notably reducing planning-time inference cost at long prediction horizons.
