CORTEXA
← Browse

Andrew G. Moore

1 paper indexed

arxivcs.LGcond-mat.dis-nncs.AI2026-06-30

Scaling Up Thermodynamic AI Models

Andrew G. Moore

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…

View free PDFSource page