Thermodynamic Computing

Ask about this topic

Overview

Thermodynamic computing studies how systems that evolve through thermal or other stochastic dynamics can solve computational tasks, including inference, sampling, and optimization. Common approaches encode problems in energy landscapes or interacting Ising variables, then use Langevin dynamics, equilibration, or reservoir networks; training methods include backpropagation and local rules such as equilibrium propagation. Demonstrations span classification and generative models, with work exploring both digital implementations and analog physical substrates. A central open tension is whether projected energy savings survive realistic accounting for control, readout, training, and heat, while maintaining useful accuracy and speed.

Latest papers

CardsList

12 papers0 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

  1. Online local learning for generative thermodynamic computing

    Sep 14, 2026Huilin Wang, Weibing DengThermodynamic ComputingLocal Learning

    Key Laboratory of Quark and Lepton Physics (MOE) and Institute of Particle Physics, Central China Normal University, Wuhan 430079, China

  2. The thermodynamic freedom of a thermodynamic computer

    Aug 28, 2026Stephen WhitelamEfficient InferenceThermodynamic Computing

    Molecular Foundry, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA

  3. Thermalizing Stochastic Programs

    Aug 3, 2026Mirko Amico, Andraž Jelinčič, Colin Oscar Nancarrow +6Thermodynamic ComputingEnergy-Efficient ML

    Extropic Corporation, San Francisco, California 94111, USA · CUNY Graduate Center, New York, NY, USA

  4. CN101 - A Digital Thermodynamic Computer for Generative AI

    Aug 1, 2026Lars Holdijk, Denis Melanson, Zier Mensch +14Thermodynamic Computing

    Normal Computing Corporation, New York, USA · University of Oxford, Oxford, United Kingdom · University of Amsterdam, Amsterdam, Netherlands +2

  5. A Blueprint for Equilibrium-Based Differentiable Continuous-Variable Thermodynamic Computing

    Jul 17, 2026Owen Lockwood, Jérémy Béjanin, Joost Bus +4Thermodynamic ComputingEnergy-Efficient ML

    Extropic Corporation, San Francisco, California 94111, USA · 2Noumenal Labs Inc, Dallas, Texas 75229, USA · 1Extropic Corporation, San Francisco, California 94111, USA +1

  6. Moment-Resolved Readout and Reservoir Diversity in Nonequilibrium Langevin Computing

    Jul 16, 2026JiZheng Duan, MingYang Zhao, YanWei Chen +1Thermodynamic ComputingReservoir Computing

    Institute of Modern Physics, Chinese Academy of Sciences

  7. Scaling Up Thermodynamic AI Models

    Jun 30, 2026Andrew G. MooreEfficient Neural Network InferenceAI Accelerator Inference

  8. Energy-efficient codon optimization on thermodynamic hardware

    Jun 15, 2026Andraz Jelincic, Ross C. WalkerThermodynamic ComputingIsing Model

    1Extropic Corp., Boston, MA

  9. Thermodynamic Networks: Harnessing Non-Equilibrium Steady States for Computation

    May 15, 2026Patryk Lipka-Bartosik, Gianmichele Blasi, Javier Lalueza Puértolas +3Thermodynamic ComputingFunction Approximation

    Center for Theoretical Physics, Polish Academy of Sciences, Warsaw, Poland · Instituto de Física Interdisciplinar y Sistemas Complejos IFISC (CSIC-UIB), E-07122 Palma de Mallorca, Spain · Física Teòrica: Informació i Fenòmens Quàntics, Department de Física, Universitat Autònoma de Barcelona, 08193 Bellaterra (Barcelona), Spain +1

  10. Symmetric Equilibrium Propagation for Thermodynamic Diffusion Training

    Apr 26, 2026Aditi DeEquilibrium PropagationThermodynamic Computing

  11. Thermodynamic Diffusion Inference with Minimal Digital Conditioning

    Apr 15, 2026Aditi DeAnalog ComputingDiffusion Models

  12. Evolutionary design of thermodynamic logic gates and their heat emission

    Feb 13, 2026Stephen WhitelamThermodynamic Computing

    Molecular Foundry, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA