用迭代推理实现小模型高效最优控制，毫秒级响应且内存占用极低。
Tiny Recursive Control: Iterative Reasoning for Efficient Optimal Control
- 通过重复使用小型网络的递归结构，以深度换容量，不增加参数量。
- 在振子稳定与燃料约束着陆任务中，控制成本接近最优，推理仅需毫秒。
- 适合资源受限的嵌入式航空航天系统，比大语言模型节省近百倍内存。
神经网络控制器参数量动辄数百万甚至数十亿，难以满足嵌入式航空航天系统对功耗和延迟的严苛要求。本文提出微型递归控制（TRC），基于一个反直觉原理：容量可来自迭代深度而非参数数量。TRC采用约150万参数的紧凑网络，通过两级分层潜在结构反复执行，模拟轨迹并根据跟踪误差修正控制序列。由于所有修正步骤共享相同权重，增加迭代次数仅提升计算量而不增加内存。我们在非线性控制问题上评估TRC，包括振子稳定与燃料约束着陆。结果表明，TRC在多个任务中达到近最优控制代价，单次推理在GPU上仅需毫秒级时间，内存占用低于10MB，较语言模型基线减少两个数量级。这证明递归推理从离散任务成功迁移至连续控制合成。
原文摘要 · Abstract (English)
Neural network controllers increasingly demand millions of parameters, and language model approaches push into the billions. For embedded aerospace systems with strict power and latency constraints, this scaling is prohibitive. We present Tiny Recursive Control (TRC), a neural architecture based on a counterintuitive principle: capacity can emerge from iteration depth rather than parameter count. TRC applies compact networks (approximately 1.5M parameters) repeatedly through a two-level hierarchical latent structure, refining control sequences by simulating trajectories and correcting based on tracking error. Because the same weights process every refinement step, adding iterations increases computation without increasing memory. We evaluate TRC on nonlinear control problems including oscillator stabilization and powered descent with fuel constraints. Across these domains, TRC achieves near-optimal control costs while requiring only millisecond-scale inference on GPU and under 10~MB memory, two orders of magnitude smaller than language model baselines. These results demonstrate that recursive reasoning, previously confined to discrete tasks, transfers effectively to continuous control synthesis.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。