Resistive memory-based zero-shot liquid state machine for multimodal event data learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Ning, Wang, Shaocong, Li, Yi, Wang, Bo, Shi, Shuhui, He, Yangu, Zhang, Woyu, Yu, Yifei, Zhang, Yue, Zhang, Xinyuan, Wong, Kwunhang, Wang, Songqi, Chen, Xiaoming, Jiang, Hao, Zhang, Xumeng, Lin, Peng, Xu, Xiaoxin, Qi, Xiaojuan, Wang, Zhongrui, Shang, Dashan, Liu, Qi, Liu, Ming
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929668892917760
author Lin, Ning
Wang, Shaocong
Li, Yi
Wang, Bo
Shi, Shuhui
He, Yangu
Zhang, Woyu
Yu, Yifei
Zhang, Yue
Zhang, Xinyuan
Wong, Kwunhang
Wang, Songqi
Chen, Xiaoming
Jiang, Hao
Zhang, Xumeng
Lin, Peng
Xu, Xiaoxin
Qi, Xiaojuan
Wang, Zhongrui
Shang, Dashan
Liu, Qi
Liu, Ming
author_facet Lin, Ning
Wang, Shaocong
Li, Yi
Wang, Bo
Shi, Shuhui
He, Yangu
Zhang, Woyu
Yu, Yifei
Zhang, Yue
Zhang, Xinyuan
Wong, Kwunhang
Wang, Songqi
Chen, Xiaoming
Jiang, Hao
Zhang, Xumeng
Lin, Peng
Xu, Xiaoxin
Qi, Xiaojuan
Wang, Zhongrui
Shang, Dashan
Liu, Qi
Liu, Ming
contents The human brain is a complex spiking neural network (SNN), capable of learning multimodal signals in a zero-shot manner by generalizing existing knowledge. Remarkably, it maintains minimal power consumption through event-based signal propagation. However, replicating the human brain in neuromorphic hardware presents both hardware and software challenges. Hardware limitations, such as the slowdown of Moore's law and Von Neumann bottleneck, hinder the efficiency of digital computers. Additionally, SNNs are characterized by their software training complexities. To this end, we propose a hardware-software co-design on a 40 nm 256 Kb in-memory computing macro that physically integrates a fixed and random liquid state machine (LSM) SNN encoder with trainable artificial neural network (ANN) projections. We showcase the zero-shot LSM-based learning of multimodal events on the N-MNIST and N-TIDIGITS datasets, including visual and audio data association, as well as neural and visual data alignment for brain-machine interfaces. Our co-design achieves classification accuracy comparable to fully optimized software models, resulting in a 152.83 and 393.07-fold reduction in training costs compared to SOTA contrastive language-image pre-training (CLIP) and Prototypical networks, and a 23.34 and 160-fold improvement in energy efficiency compared to cutting-edge digital hardware, respectively. These proof-of-principle prototypes demonstrate zero-shot multimodal events learning capability for emerging efficient and compact neuromorphic hardware.
format Preprint
id arxiv_https___arxiv_org_abs_2307_00771
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Resistive memory-based zero-shot liquid state machine for multimodal event data learning
Lin, Ning
Wang, Shaocong
Li, Yi
Wang, Bo
Shi, Shuhui
He, Yangu
Zhang, Woyu
Yu, Yifei
Zhang, Yue
Zhang, Xinyuan
Wong, Kwunhang
Wang, Songqi
Chen, Xiaoming
Jiang, Hao
Zhang, Xumeng
Lin, Peng
Xu, Xiaoxin
Qi, Xiaojuan
Wang, Zhongrui
Shang, Dashan
Liu, Qi
Liu, Ming
Emerging Technologies
The human brain is a complex spiking neural network (SNN), capable of learning multimodal signals in a zero-shot manner by generalizing existing knowledge. Remarkably, it maintains minimal power consumption through event-based signal propagation. However, replicating the human brain in neuromorphic hardware presents both hardware and software challenges. Hardware limitations, such as the slowdown of Moore's law and Von Neumann bottleneck, hinder the efficiency of digital computers. Additionally, SNNs are characterized by their software training complexities. To this end, we propose a hardware-software co-design on a 40 nm 256 Kb in-memory computing macro that physically integrates a fixed and random liquid state machine (LSM) SNN encoder with trainable artificial neural network (ANN) projections. We showcase the zero-shot LSM-based learning of multimodal events on the N-MNIST and N-TIDIGITS datasets, including visual and audio data association, as well as neural and visual data alignment for brain-machine interfaces. Our co-design achieves classification accuracy comparable to fully optimized software models, resulting in a 152.83 and 393.07-fold reduction in training costs compared to SOTA contrastive language-image pre-training (CLIP) and Prototypical networks, and a 23.34 and 160-fold improvement in energy efficiency compared to cutting-edge digital hardware, respectively. These proof-of-principle prototypes demonstrate zero-shot multimodal events learning capability for emerging efficient and compact neuromorphic hardware.
title Resistive memory-based zero-shot liquid state machine for multimodal event data learning
topic Emerging Technologies
url https://arxiv.org/abs/2307.00771