Neuromorphic Overparameterisation and Few-Shot Learning in Multilayer Physical Neural Networks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912040540438528 |
|---|---|
| author | Stenning, Kilian D. Gartside, Jack C. Manneschi, Luca Cheung, Christopher T. S. Chen, Tony Vanstone, Alex Love, Jake Holder, Holly H. Caravelli, Francesco Kurebayashi, Hidekazu Everschor-Sitte, Karin Vasilaki, Eleni Branford, Will R. |
| author_facet | Stenning, Kilian D. Gartside, Jack C. Manneschi, Luca Cheung, Christopher T. S. Chen, Tony Vanstone, Alex Love, Jake Holder, Holly H. Caravelli, Francesco Kurebayashi, Hidekazu Everschor-Sitte, Karin Vasilaki, Eleni Branford, Will R. |
| contents | Physical neuromorphic computing, exploiting the complex dynamics of physical systems, has seen rapid advancements in sophistication and performance. Physical reservoir computing, a subset of neuromorphic computing, faces limitations due to its reliance on single systems. This constrains output dimensionality and dynamic range, limiting performance to a narrow range of tasks. Here, we engineer a suite of nanomagnetic array physical reservoirs and interconnect them in parallel and series to create a multilayer neural network architecture. The output of one reservoir is recorded, scaled and virtually fed as input to the next reservoir. This networked approach increases output dimensionality, internal dynamics and computational performance. We demonstrate that a physical neuromorphic system can achieve an overparameterised state, facilitating meta-learning on small training sets and yielding strong performance across a wide range of tasks. Our approach's efficacy is further demonstrated through few-shot learning, where the system rapidly adapts to new tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2211_06373 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Neuromorphic Overparameterisation and Few-Shot Learning in Multilayer Physical Neural Networks Stenning, Kilian D. Gartside, Jack C. Manneschi, Luca Cheung, Christopher T. S. Chen, Tony Vanstone, Alex Love, Jake Holder, Holly H. Caravelli, Francesco Kurebayashi, Hidekazu Everschor-Sitte, Karin Vasilaki, Eleni Branford, Will R. Mesoscale and Nanoscale Physics Physical neuromorphic computing, exploiting the complex dynamics of physical systems, has seen rapid advancements in sophistication and performance. Physical reservoir computing, a subset of neuromorphic computing, faces limitations due to its reliance on single systems. This constrains output dimensionality and dynamic range, limiting performance to a narrow range of tasks. Here, we engineer a suite of nanomagnetic array physical reservoirs and interconnect them in parallel and series to create a multilayer neural network architecture. The output of one reservoir is recorded, scaled and virtually fed as input to the next reservoir. This networked approach increases output dimensionality, internal dynamics and computational performance. We demonstrate that a physical neuromorphic system can achieve an overparameterised state, facilitating meta-learning on small training sets and yielding strong performance across a wide range of tasks. Our approach's efficacy is further demonstrated through few-shot learning, where the system rapidly adapts to new tasks. |
| title | Neuromorphic Overparameterisation and Few-Shot Learning in Multilayer Physical Neural Networks |
| topic | Mesoscale and Nanoscale Physics |
| url | https://arxiv.org/abs/2211.06373 |