{"ID":2829601,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.12428","arxiv_id":"2512.12428","title":"Learning Dynamics in Memristor-Based Equilibrium Propagation","abstract":"Memristor-based in-memory computing has emerged as a promising paradigm to overcome the constraints of the von Neumann bottleneck and the memory wall by enabling fully parallelisable and energy-efficient vector-matrix multiplications. We investigate the effect of nonlinear, memristor-driven weight updates on the convergence behaviour of neural networks trained with equilibrium propagation (EqProp). Six memristor models were characterised by their voltage-current hysteresis and integrated into the EBANA framework for evaluation on two benchmark classification tasks. EqProp can achieve robust convergence under nonlinear weight updates, provided that memristors exhibit a sufficiently wide resistance range of at least an order of magnitude.","short_abstract":"Memristor-based in-memory computing has emerged as a promising paradigm to overcome the constraints of the von Neumann bottleneck and the memory wall by enabling fully parallelisable and energy-efficient vector-matrix multiplications. We investigate the effect of nonlinear, memristor-driven weight updates on the conver...","url_abs":"https://arxiv.org/abs/2512.12428","url_pdf":"https://arxiv.org/pdf/2512.12428v1","authors":"[\"Michael Döll\",\"Andreas Müller\",\"Bernd Ulmann\"]","published":"2025-12-13T18:57:05Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.ET\",\"cs.NE\"]","methods":"[]","has_code":false}