Dune:VLAのテスト時適応のための時間条件付き履歴誤差回帰を用いたゼロ次最適化器
Dune: A Zeroth-Order Optimizer with Time-Conditioned Historical Error Regression for VLA Test-Time Adaptation
VLAモデルのテスト時適応において、過去の最適化履歴から誤差を予測してゼロ次勾配推定と融合することで、逆伝播なし・低メモリで長期的な適応を安定化させる最適化手法を提案した。
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: Chengyang Dong, Ze Chen, Kai Ye, Chutian Wang, Heng Xu, Wei Mo, Maomao Li, Jiayan Xu, Jiayuan Wu, Yihao Xiao, Linhua Ye, Zhonghua Peng, Ruize Han, Jialong Sun
分類: cs.RO, cs.AI
原文アブストラクト
Test-Time Adaptation (TTA) enables Vision-Language-Action (VLA) models to handle distribution shifts during deployment without source-domain retraining. However, backpropagation-based adaptation incurs heavy memory overhead on resource-constrained robots, whereas zeroth-order (ZO) methods suffer from high estimation variance under limited perturbation budgets. We observe that while ZO estimation enables low-memory adaptation with rapid initial gains, it often stalls or degrades over long-horizon updates due to noisy gradient estimates. To address these limitations, we propose Dune, a zeroth-order optimizer that stabilizes long-horizon adaptation by coupling online historical error prediction with current perturbation observations. Specifically, Dune learns a lightweight, time-conditioned error predictor from past optimization dynamics and fuses its predictions with instantaneous ZO estimates, achieving effective variance reduction without backpropagation or extra forward passes. Additionally, we theoretically characterize how estimation variance restricts convergence in high-dimensional, small-population regimes, motivating the use of historical statistics. Extensive experiments on LIBERO and real-world robots demonstrate that Dune achieves superior success rates under a low memory footprint, with ablations confirming improved closed-loop stability and post-disturbance recovery.
