Reservoir Computation with Networks of Differentiating Neuron Ring Oscillators
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909710178844672 |
|---|---|
| author | Yeung, Alexander DelMastro, Peter Karuvally, Arjun Siegelmann, Hava Rietman, Edward Hazan, Hananel |
| author_facet | Yeung, Alexander DelMastro, Peter Karuvally, Arjun Siegelmann, Hava Rietman, Edward Hazan, Hananel |
| contents | Reservoir Computing is a machine learning approach that uses the rich repertoire of complex system dynamics for function approximation. Current approaches to reservoir computing use a network of coupled integrating neurons that require a steady current to maintain activity. Here, we introduce a small world graph of differentiating neurons that are active only when there are changes in input as an alternative to integrating neurons as a reservoir computing substrate. We find the coupling strength and network topology that enable these small world networks to function as an effective reservoir. We demonstrate the efficacy of these networks in the MNIST digit recognition task, achieving comparable performance of 90.65% to existing reservoir computing approaches. The findings suggest that differentiating neurons can be a potential alternative to integrating neurons and can provide a sustainable future alternative for power-hungry AI applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_21377 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Reservoir Computation with Networks of Differentiating Neuron Ring Oscillators Yeung, Alexander DelMastro, Peter Karuvally, Arjun Siegelmann, Hava Rietman, Edward Hazan, Hananel Neural and Evolutionary Computing Machine Learning Reservoir Computing is a machine learning approach that uses the rich repertoire of complex system dynamics for function approximation. Current approaches to reservoir computing use a network of coupled integrating neurons that require a steady current to maintain activity. Here, we introduce a small world graph of differentiating neurons that are active only when there are changes in input as an alternative to integrating neurons as a reservoir computing substrate. We find the coupling strength and network topology that enable these small world networks to function as an effective reservoir. We demonstrate the efficacy of these networks in the MNIST digit recognition task, achieving comparable performance of 90.65% to existing reservoir computing approaches. The findings suggest that differentiating neurons can be a potential alternative to integrating neurons and can provide a sustainable future alternative for power-hungry AI applications. |
| title | Reservoir Computation with Networks of Differentiating Neuron Ring Oscillators |
| topic | Neural and Evolutionary Computing Machine Learning |
| url | https://arxiv.org/abs/2507.21377 |