AI & ML
Linear Recurrent Memory Suffices to Distil a World-Model Policy for Robot Air Hockey
arXiv cs.RO0 views
arXiv:2609.39151v2 Announce Type: replace Abstract: Does memory-dependent control need nonlinear recurrent dynamics? We study simulated air-hockey defence under temporary loss of puck tracking. A DreamerV3 teacher outperforms a memoryless policy under tracking loss, and erasing its recurrent state at the moment tracking fails cuts its save rate sharply, so the task requires memory. We distil this teacher into compact recurrent policies with a 64-dimensional state, combining a diagonal linear recurrence with an optional rank-$k$ nonlinear innovation while the observation encoder and action head stay nonlinear. Across five matched seeds, the purely linear recurrent model ($k=0$) matches both the GRU baseline and the teacher throughout the tested range of tracking loss. Increasing nonlinear innovation rank provides no measured benefit. This result comes from a fresh test split that was opened only after all models and analyses were frozen. Under 5\,mm observation noise, without retraining, the linear student keeps a higher save rate than the GRU baseline, while using fewer recurrent parameters and less computation. These results suggest that, for this memory-dependent control task, nonlinear representation learning around a simple linear memory can be sufficient, and that nonlinear recurrent dynamics are not necessarily required.
Source:https://arxiv.org/abs/2609.39151PreviousOne from Infinity: Actualizing Futures from Pretrained World Models into Robot ActionsNextVLALagBench: Understanding When and Why VLA Inference Latency Matters
Related posts
SALA: Semantically Aligned Latent Actions Enable Dual-Axis Scaling of Embodied Foundation Models
0arXiv cs.RO
What Can a VLA Learn from the Agent That Calls It? Task-Level Experience Distillation for Embodied Navigation
0arXiv cs.RO
Fast Inference-Time Policy Refinement via Action-Affine World Models
0arXiv cs.RO
Data-Driven Configuration of Protective Separation Zones in Collaborative Robot Cells from Operator Occupancy
0arXiv cs.RO
VLALagBench: Understanding When and Why VLA Inference Latency Matters
0arXiv cs.RO
One from Infinity: Actualizing Futures from Pretrained World Models into Robot Actions
0arXiv cs.RO