In-Hardware Learning of Multilayer Spiking Neural Networks on a Neuromorphic Processor

Although widely used in machine learning, backpropagation cannot directly be applied to SNN training and is not feasible on a neuromorphic processor that emulates biological neuron and synapses. This work presents a spike-based backpropagation algorithm with biological plausible local update rules a...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2021 58th ACM/IEEE Design Automation Conference (DAC) s. 367 - 372
Hlavní autoři: Shrestha, Amar, Fang, Haowen, Rider, Daniel Patrick, Mei, Zaidao, Qiu, Qinru
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 05.12.2021
Témata:
Backpropagation algorithms
Bio-Inspired Approaches
Biology
Hardware
Neuromorphic Computing
Neuromorphics
Nonhomogeneous media
Spiking Neural Networks
Supervised learning
Training
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Although widely used in machine learning, backpropagation cannot directly be applied to SNN training and is not feasible on a neuromorphic processor that emulates biological neuron and synapses. This work presents a spike-based backpropagation algorithm with biological plausible local update rules and adapts it to fit the constraint in a neuromorphic hardware. The algorithm is implemented on Intel's Loihi chip enabling low power in-hardware supervised online learning of multilayered SNNs for mobile applications. We test this implementation on MNIST, Fashion-MNIST, CIFAR-10 and MSTAR datasets with promising performance and energy-efficiency, and demonstrate a possibility of incremental online learning with the implementation.
DOI:10.1109/DAC18074.2021.9586323