Thermodynamic Networks: Harnessing Non-Equilibrium Steady States for Computation

2026-05-15

AI Summary

This work proposes a physical computing paradigm grounded in nonequilibrium steady states, wherein conserved quantities are exchanged between finite reservoirs through a thermodynamic network that relaxes to encode computational solutions. The key insight is that negative differential conductance (NDC) serves as the critical physical mechanism governing computational expressivity: networks exhibiting NDC achieve universal function approximation. By integrating nonequilibrium thermodynamic modeling, the approach leverages intrinsic relaxation dynamics for training, implemented on platforms such as quantum dots and enzymatic reaction networks. Experimental results demonstrate strong performance on tasks including sine function fitting and MNIST classification, confirming the framework’s efficiency, autonomy, and universality.

Abstract

We introduce thermodynamic networks, a general framework for autonomous, physics-based computation using non-equilibrium steady states. These networks are modeled as a collection of finite-size reservoirs that exchange conserved quantities--such as electric charge or molecular number--while relaxing to a non-equilibrium steady state, which encodes the solution of a computational problem. We identify Negative Differential Conductance (NDC) as the critical physical property governing the computational expressivity of the thermodynamic network. While networks lacking NDC are restricted to computing monotonic functions, the presence of NDC enables universal function approximation. For the training of the network, we use protocols that take advantage of the natural tendency of the system to equilibrate. We illustrate the versatility of our approach via two different platforms: quantum dot networks and enzymatic reaction networks. Both systems can be engineered to have NDC, enabling high performance in standard benchmarks, including sine function approximation and MNIST digit classification. Overall, our work establishes a rigorous link between non-equilibrium steady states and computational expressivity.

Problem

• thermodynamic networks
• non-equilibrium steady states
• computational expressivity
• Negative Differential Conductance
• autonomous computation

Innovation

• Thermodynamic Networks
• Non-Equilibrium Steady States
• Negative Differential Conductance
• Physics-Based Computation
• Universal Function Approximation