{"ID":5552826,"CreatedAt":"2026-07-02T01:54:51.863792489Z","UpdatedAt":"2026-07-03T20:14:26.82372516Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.00170","arxiv_id":"2607.00170","title":"Scaling Up Thermodynamic AI Models","abstract":"Thermodynamic computing devices based on the Ising model show great promise for low-power AI inference and edge computing, but scalable methods for training large models for such hardware remain limited. Prior theory shows that the time-averaged behavior of high-temperature Gibbs-sampled Ising systems can implement feed-forward neural inference. We turn this theoretical correspondence into a scalable and purely backpropagation-based algorithm for training deep convolutional networks for thermodynamic inference on Ising machine hardware. Our image classification models achieve accuracies of 94.9% on CIFAR-10 and 76.0% on CIFAR-100 under binary Gibbs sampling. We then develop and experimentally validate a mathematical theory relating inference cost to accuracy and controlling autocorrelation times. Subsequently, we calculate asymptotic results showing that inference cost is bounded by a well-controlled tradeoff with performance and exhibit algorithms for computing optimal inference schedules. Finally, we discuss implications for hardware development and the future of high-temperature thermodynamic AI models.","short_abstract":"Thermodynamic computing devices based on the Ising model show great promise for low-power AI inference and edge computing, but scalable methods for training large models for such hardware remain limited. Prior theory shows that the time-averaged behavior of high-temperature Gibbs-sampled Ising systems can implement fee...","url_abs":"https://arxiv.org/abs/2607.00170","url_pdf":"https://arxiv.org/pdf/2607.00170v1","authors":"[\"Andrew G. Moore\"]","published":"2026-06-30T20:44:57Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cond-mat.dis-nn\",\"cs.AI\"]","methods":"[]","has_code":false}