Low-power Spike-based Wearable Analytics on RRAM Crossbars
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929707465834496 |
|---|---|
| author | Bhattacharjee, Abhiroop Shi, Jinquan Chen, Wei-Chen Wang, Xinxin Panda, Priyadarshini |
| author_facet | Bhattacharjee, Abhiroop Shi, Jinquan Chen, Wei-Chen Wang, Xinxin Panda, Priyadarshini |
| contents | This work introduces a spike-based wearable analytics system utilizing Spiking Neural Networks (SNNs) deployed on an In-memory Computing engine based on RRAM crossbars, which are known for their compactness and energy-efficiency. Given the hardware constraints and noise characteristics of the underlying RRAM crossbars, we propose online adaptation of pre-trained SNNs in real-time using Direct Feedback Alignment (DFA) against traditional backpropagation (BP). Direct Feedback Alignment (DFA) learning, that allows layer-parallel gradient computations, acts as a fast, energy & area-efficient method for online adaptation of SNNs on RRAM crossbars, unleashing better algorithmic performance against those adapted using BP. Through extensive simulations using our in-house hardware evaluation engine called DFA_Sim, we find that DFA achieves upto 64.1% lower energy consumption, 10.1% lower area overhead, and a 2.1x reduction in latency compared to BP, while delivering upto 7.55% higher inference accuracy on human activity recognition (HAR) tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_06736 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Low-power Spike-based Wearable Analytics on RRAM Crossbars Bhattacharjee, Abhiroop Shi, Jinquan Chen, Wei-Chen Wang, Xinxin Panda, Priyadarshini Emerging Technologies Artificial Intelligence Hardware Architecture This work introduces a spike-based wearable analytics system utilizing Spiking Neural Networks (SNNs) deployed on an In-memory Computing engine based on RRAM crossbars, which are known for their compactness and energy-efficiency. Given the hardware constraints and noise characteristics of the underlying RRAM crossbars, we propose online adaptation of pre-trained SNNs in real-time using Direct Feedback Alignment (DFA) against traditional backpropagation (BP). Direct Feedback Alignment (DFA) learning, that allows layer-parallel gradient computations, acts as a fast, energy & area-efficient method for online adaptation of SNNs on RRAM crossbars, unleashing better algorithmic performance against those adapted using BP. Through extensive simulations using our in-house hardware evaluation engine called DFA_Sim, we find that DFA achieves upto 64.1% lower energy consumption, 10.1% lower area overhead, and a 2.1x reduction in latency compared to BP, while delivering upto 7.55% higher inference accuracy on human activity recognition (HAR) tasks. |
| title | Low-power Spike-based Wearable Analytics on RRAM Crossbars |
| topic | Emerging Technologies Artificial Intelligence Hardware Architecture |
| url | https://arxiv.org/abs/2502.06736 |