Tiny Recursive Control: Iterative Reasoning for Efficient Optimal Control
- URL: http://arxiv.org/abs/2512.16824v1
- Date: Thu, 18 Dec 2025 18:05:05 GMT
- Title: Tiny Recursive Control: Iterative Reasoning for Efficient Optimal Control
- Authors: Amit Jain, Richard Linares,
- Abstract summary: We present Tiny Recursive Control (TRC), a neural architecture based on a counterintuitive iteration principle.<n>TRC applies compact networks (approximately 1.5M parameters) repeatedly through a two-level hierarchical latent structure.<n>We show TRC achieves near-optimal control costs while requiring only millisecond-scale inference on GPU and under 10MB memory.
- Score: 2.258690092379457
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: 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.
Related papers
- Joint Geometric and Trajectory Consistency Learning for One-Step Real-World Super-Resolution [14.52346301984322]
Diffusion-based Real-World Image Super-Resolution (Real-ISR) achieves impressive perceptual quality but suffers from high computational costs due to iterative sampling.<n>We propose GTASR (Geometric Trajectory Alignment Super-Resolution), a simple yet effective consistency training paradigm for Real-ISR.
arXiv Detail & Related papers (2026-02-27T18:13:31Z) - Intelligent Control of Collisional Architectures for Deterministic Multipartite State Engineering [0.0]
We introduce an intelligent, constraint-aware control framework for deterministic generation of symmetric Dicke states $|D_n(m)rangle$ in repeated excitation-within-interaction architectures.<n>The protocol employs partialSWAP collisions between two disjoint qubit registers, mediated by $m$ ancillary shuttle'' qubits, and poses Dickestate preparation as a emph-loop design problem.
arXiv Detail & Related papers (2026-02-09T11:15:32Z) - RaBiT: Residual-Aware Binarization Training for Accurate and Efficient LLMs [5.782015253162346]
Residual binarization enables matmul-free inference by stacking binary layers.<n>We propose RaBiT, a novel quantization framework that resolves coadaptation by algorithmically enforcing a residual hierarchy.<n>RaBiT achieves state-of-the-art performance, rivals even hardware-intensive Vector Quantization (VQ) methods, and delivers a $4.49times$ inference speed-up over full-precision models.
arXiv Detail & Related papers (2026-02-05T06:41:11Z) - TNT: Improving Chunkwise Training for Test-Time Memorization [62.78875147721906]
Recurrent neural networks (RNNs) with deep test-time memorization modules, such as Titans and TTT, represent a promising, linearly-scaling paradigm distinct from Transformers.<n>We introduce TNT, a novel training paradigm that decouples training efficiency from inference performance through a two-stage process.<n>TNT achieves a substantial acceleration in training speed-up to 17 times faster than the most accurate baseline configuration.
arXiv Detail & Related papers (2025-11-10T17:45:09Z) - PT$^2$-LLM: Post-Training Ternarization for Large Language Models [52.4629647715623]
Large Language Models (LLMs) have shown impressive capabilities across diverse tasks, but their large memory and compute demands hinder deployment.<n>We propose PT$2$-LLM, a post-training ternarization framework tailored for LLMs.<n>At its core is an Asymmetric Ternary Quantizer equipped with a two-stage refinement pipeline.
arXiv Detail & Related papers (2025-09-27T03:01:48Z) - Comba: Improving Bilinear RNNs with Closed-loop Control [57.800320390698516]
We introduce the concept of Bilinear RNNs with a comprehensive analysis on the advantages and limitations of these models.<n>We propose a novel Bilinear RNN variant named Comba, which adopts a scalar-plus-low-rank state transition, with both state feedback and output feedback corrections.<n>We also implement a hardware-efficient chunk-wise parallel kernel in Triton and train models with 340M/1.3B parameters on large-scale corpus.
arXiv Detail & Related papers (2025-06-03T05:44:50Z) - MOGNET: A Mux-residual quantized Network leveraging Online-Generated weights [2.7036595757881323]
MOGNET is a compact model architecture compatible with resource-limited hardware.<n>It can achieve higher accuracy with a clear gap up to 1% at a similar or even lower model size.
arXiv Detail & Related papers (2025-01-16T13:30:20Z) - Neural Operators for Boundary Stabilization of Stop-and-go Traffic [1.90298817989995]
This paper introduces a novel approach to PDE boundary control design using neural operators.
We present two distinct neural operator learning schemes aimed at stabilizing the traffic PDE system.
It is proved that the NO-based closed-loop system is practical stable under certain approximation accuracy conditions in NO-learning.
arXiv Detail & Related papers (2023-12-16T08:18:39Z) - CACTO: Continuous Actor-Critic with Trajectory Optimization -- Towards
global optimality [5.0915256711576475]
This paper presents a novel algorithm for the continuous control of dynamical systems that combines Trayy (TO) and Reinforcement Learning (RL) in a single trajectory.
arXiv Detail & Related papers (2022-11-12T10:16:35Z) - Efficient Micro-Structured Weight Unification and Pruning for Neural
Network Compression [56.83861738731913]
Deep Neural Network (DNN) models are essential for practical applications, especially for resource limited devices.
Previous unstructured or structured weight pruning methods can hardly truly accelerate inference.
We propose a generalized weight unification framework at a hardware compatible micro-structured level to achieve high amount of compression and acceleration.
arXiv Detail & Related papers (2021-06-15T17:22:59Z) - Edge Federated Learning Via Unit-Modulus Over-The-Air Computation
(Extended Version) [64.76619508293966]
This paper proposes a unit-modulus over-the-air computation (UM-AirComp) framework to facilitate efficient edge federated learning.
It uploads simultaneously local model parameters and updates global model parameters via analog beamforming.
We demonstrate the implementation of UM-AirComp in a vehicle-to-everything autonomous driving simulation platform.
arXiv Detail & Related papers (2021-01-28T15:10:22Z) - High-bandwidth nonlinear control for soft actuators with recursive
network models [1.4174475093445231]
We present a high-bandwidth, lightweight, and nonlinear output tracking technique for soft actuators using Newton-Raphson.
This technique allows for reduced model sizes and increased control loop frequencies when compared with conventional RNN models.
arXiv Detail & Related papers (2021-01-04T18:12:41Z) - Adaptive Subcarrier, Parameter, and Power Allocation for Partitioned
Edge Learning Over Broadband Channels [69.18343801164741]
partitioned edge learning (PARTEL) implements parameter-server training, a well known distributed learning method, in wireless network.
We consider the case of deep neural network (DNN) models which can be trained using PARTEL by introducing some auxiliary variables.
arXiv Detail & Related papers (2020-10-08T15:27:50Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.