具有混合密度头的液体网络用于高效模仿学习
Liquid Networks with Mixture Density Heads for Efficient Imitation Learning
AI总结：
本文比较了具有混合密度头的液体网络与扩散策略在多个任务中的性能，发现液体网络参数更少、预测误差更低且推理更快，且在样本效率实验中表现更稳健。
摘要：
我们比较了具有混合密度头的液体神经网络与扩散策略在Push-T、RoboMimic Can和PointMaze任务中的表现，采用共享骨干的比较协议以隔离策略头的影响。在不同任务中，液体策略使用约一半的参数（4.3M vs. 8.6M），预测误差低2.4倍，推理速度快1.8倍。在样本效率实验中，液体模型在低数据和中等数据情况下表现更稳健。Push-T和PointMaze的闭环结果与离线排名方向一致但更嘈杂，表明强大的离线密度建模有助于部署，但不完全决定闭环成功。总体而言，液体递归多模态策略为模仿学习提供了一种紧凑且实用的替代方案，替代迭代去噪方法。
英文摘要：
We compare liquid neural networks with mixture density heads against diffusion policies on Push-T, RoboMimic Can, and PointMaze under a shared-backbone comparison protocol that isolates policy-head effects under matched inputs, training budgets, and evaluation settings. Across tasks, liquid policies use roughly half the parameters (4.3M vs. 8.6M), achieve 2.4x lower offline prediction error, and run 1.8 faster at inference. In sample-efficiency experiments spanning 1% to 46.42% of training data, liquid models remain consistently more robust, with especially large gains in low-data and medium-data regimes. Closed-loop results on Push-T and PointMaze are directionally consistent with offline rankings but noisier, indicating that strong offline density modeling helps deployment while not fully determining closed-loop success. Overall, liquid recurrent multimodal policies provide a compact and practical alternative to iterative denoising for imitation learning.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
