外观
论文arXiv 2610.0569223 天前
面向主动推断控制的生成式轨迹模型基准测试
原文链接：Benchmarking Generative Trajectory Models for Active-Inference Control（arXiv 2610.05692 · 发布日期：2026-10-05）
Benchmarking Generative Trajectory Models for Active-Inference Control
提出GenAIF用单一生成轨迹模型从演示和动作干预学习，并在MuJoCo多物理条件操控任务中基准比较多类模型。
核心要点
- 面向难以显式建模动力学的复杂系统，利用轨迹演示学习主动推断控制。
- GenAIF以一个生成轨迹模型提供目标条件策略分布和状态到观测似然映射。
- 在MuJoCo多物理条件操控任务中比较扩散、autoregressive Transformers、CVAE和flow matching。
- 扩散控制最强且倾斜变化后信念更新最快，CVAE短期预测相近且推理更快。
验证范围与局限：验证范围为MuJoCo操控任务的多种物理条件，以及回放中的未预告倾斜变化。
英文摘要 (Abstract)
Learning from trajectory demonstrations offers a route to active-inference control of complex systems whose dynamics are difficult to model explicitly. We introduce generative active-inference control (GenAIF), in which one generative trajectory model learns from demonstrations and measured action interventions to supply a goal-conditioned policy distribution and a state-to-observation likelihood mapping. From this control design, we derive three model requirements: (i) useful action proposals, (ii) accurate prediction under imposed actions, and (iii) probabilistic observation evidence for belief updating and expected information gain. We benchmark diffusion, autoregressive Transformers, conditional variational autoencoders (CVAEs), and flow matching in a MuJoCo manipulation task with multiple physical conditions. Diffusion delivers the strongest control across the tested dynamics, while CVAE combines comparable short-horizon prediction with much faster inference. Correct conditioning is decisive, and trajectory reuse offers further computational savings. In replay after an unannounced tilt change, pretrained diffusion updates belief fastest among the original models; fine-tuning on recovery demonstrations further accelerates identification and sustains accurate tracking. These findings support the use of shared generative trajectory models to connect action proposal, controlled prediction, and observation evidence within GenAIF.
作者：Yulin Li, Mohsen A. Jafari, Andrea Matta
来源
本页事实对应的公开页面，以原文为准。