SEnsitivity Modulated Importance Networking and Rehearsal for Spike Domain Incremental Learning

Zaidao Mei, Boyu Wang, Daniel Rider, Qinru Qiu

Research output: Chapter in Book/Entry/PoemConference contribution

Abstract

Incremental learning is a challenging task in the field of machine learning, and it is a key step towards autonomous learning and adaptation. With the increasing attention to neuromorphic computing, there is an urgent need to investigate incremental learning techniques that can work in this paradigm to maintain energy efficiency while benefiting from flexibility and adaptability. In this paper, we present SEMINAR (sensitivity modulated importance networking and rehearsal), an incremental learning algorithm explicitly designed for EMSTDP (Error Modulated SynapticTimingDependent Plasticity), which performs supervised learning for multi-layer spiking neural networks (SNN) implemented on neuromorphic hardware, such as Loihi. SEMINAR uses critical synapse selection, differential learning rate, and a replay buffer to enable the model to retain past knowledge while maintaining flexibility to learn new tasks. Our experimental results show that, when combined with the EM-STDP, SEMINAR outperforms different baseline incremental learning algorithms and gives more than 4% improvement on several widely used datasets such as Split-MNIST, Split-Fashion MNIST, Split-NMNIST, and MSTAR.

Original languageEnglish (US)
Title of host publicationICONS 2023 - Proceedings of International Conference on Neuromorphic Systems 2023
PublisherAssociation for Computing Machinery
ISBN (Electronic)9798400701757
DOIs
StatePublished - Aug 1 2023
Event2023 International Conference on Neuromorphic Systems, ICONS 2023 - Santa Fe, United States
Duration: Aug 1 2023Aug 3 2023

Publication series

NameACM International Conference Proceeding Series

Conference

Conference2023 International Conference on Neuromorphic Systems, ICONS 2023
Country/TerritoryUnited States
CitySanta Fe
Period8/1/238/3/23

Keywords

  • continual learning
  • neuromorphic computing
  • spiking neural network

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Fingerprint

Dive into the research topics of 'SEnsitivity Modulated Importance Networking and Rehearsal for Spike Domain Incremental Learning'. Together they form a unique fingerprint.

Cite this