Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.24113 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911543709401088 |
|---|---|
| author | Haag, Jonathan Metzner, Christian Zendrikov, Dmitrii Indiveri, Giacomo Grewe, Benjamin De Luca, Chiara Saponati, Matteo |
| author_facet | Haag, Jonathan Metzner, Christian Zendrikov, Dmitrii Indiveri, Giacomo Grewe, Benjamin De Luca, Chiara Saponati, Matteo |
| contents | On-chip learning is key to scalable and adaptive neuromorphic systems, yet existing training methods are either difficult to implement in hardware or overly restrictive. However, recent studies show that feedback-control optimizers can enable expressive, on-chip training of neuromorphic devices. In this work, we present a proof-of-concept implementation of such feedback-control optimizers on a mixed-signal neuromorphic processor. We assess the proposed approach in an In-The-Loop(ITL) training setup on both a binary classification task and the nonlinear Yin-Yang problem, demonstrating on-chip training that matches the performance of numerical simulations and gradient-based baselines. Our results highlight the feasibility of feedback-driven, online learning under realistic mixed-signal constraints, and represent a co-design approach toward embedding such rules directly in silicon for autonomous and adaptive neuromorphic computing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_24113 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Mixed-signal implementation of feedback-control optimizer for single-layer Spiking Neural Networks Haag, Jonathan Metzner, Christian Zendrikov, Dmitrii Indiveri, Giacomo Grewe, Benjamin De Luca, Chiara Saponati, Matteo Machine Learning On-chip learning is key to scalable and adaptive neuromorphic systems, yet existing training methods are either difficult to implement in hardware or overly restrictive. However, recent studies show that feedback-control optimizers can enable expressive, on-chip training of neuromorphic devices. In this work, we present a proof-of-concept implementation of such feedback-control optimizers on a mixed-signal neuromorphic processor. We assess the proposed approach in an In-The-Loop(ITL) training setup on both a binary classification task and the nonlinear Yin-Yang problem, demonstrating on-chip training that matches the performance of numerical simulations and gradient-based baselines. Our results highlight the feasibility of feedback-driven, online learning under realistic mixed-signal constraints, and represent a co-design approach toward embedding such rules directly in silicon for autonomous and adaptive neuromorphic computing. |
| title | Mixed-signal implementation of feedback-control optimizer for single-layer Spiking Neural Networks |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2603.24113 |