[论文] A Blueprint for Equilibrium-Based Differentiable Continuous-Variable Thermodynamic Computing

论文概要

研究领域: cs.LG, cs.ET, physics.app-ph
作者: Owen Lockwood, Jérémy Béjanin, Joost Bus
发布时间: 2026-07-21
arXiv: 2507.15487

中文摘要

为应对机器学习工作负载不断增长的能耗和延迟需求，我们引入了一种基于热力学的计算栈蓝图，利用物理硬件中的随机模拟过程实现高效快速计算。本工作聚焦于基于能量的热力学计算，其中随机过程由具有可调能量势的Langevin动力学精确描述。在物理硬件中实现此类势能使我们能够从基本参数化能量模型中生成和采样。我们展示了如何基于概率图模型框架，利用这些硬件原生能量模型构建和训练流行的机器学习模型类别。我们基于理论分析和数值研究，分析了该热力学范式下不同模型的运行时间和能耗。作为此类硬件的初步实验实现，我们展示了由热噪声驱动的随机模拟超导电路。这些结果共同勾勒出通向面向概率机器学习的节能热力学硬件的路径。

原文摘要

To address the escalating energy and latency demands of machine-learning workloads, we introduce a blueprint for an energy-efficient and fast thermodynamic computing stack that leverages stochastic analog processes in physical hardware. In this work, we focus on energy-based thermodynamic computing where the stochastic process is well described by Langevin dynamics with tunable energy potentials. The implementation of such potentials in physical hardware enables us to generate and sample from basic parameterized energy-based models. We demonstrate how to construct and train popular classes of machine learning models based on these hardware原生 energy-based models, using the framework of probabilistic图模型. We analyze the runtime and energy consumption of different models in this thermodynamic paradigm based on theoretical considerations and numerical studies. As a preliminary experimental realization of such hardware, we present our stochastic analog superconducting circuits driven by thermal noise. Together, these results outline a path toward energy-efficient thermodynamic hardware for probabilistic machine learning.

自动采集于 2026-07-21

#论文 #arXiv #LG #小凯