POS1 - RRAM Crossbar-Based Fault-Tolerant Binary Neural Networks (BNNs)
View/Open
PS1-4.pdf (572,4Kb) (Restricted access)
Cita com:
hdl:2117/372113
Document typeConference report
Defense date2022-05
Rights accessRestricted access - publisher's policy
Except where otherwise noted, content on this work
is licensed under a Creative Commons license
:
Attribution-NonCommercial-NoDerivs 4.0 International
Abstract
Computation-In Memory (CIM) using RRAM crossbar
array is a promising solution to realize energy-efficient
neuromorphic hardware, such as Binary Neural Networks (BNNs).
However, RRAM faults restrict the applicability of CIM for BNN
implementation. To address this issue, we propose a fault tolerance
framework to mitigate the impact of RRAM faults on the accuracy
of CIM-based BNN hardware. Evaluation results using MNIST,
Fashion-MNIST and CIFAR-10 datasets demonstrate that the
proposed framework outperforms the related works as it restores
more than 99% of the RRAM fault induced accuracy reduction
with relatively less overhead.
CitationGebregiorgis, A.; Zografou, A.; Hamdioui, S. POS1 - RRAM Crossbar-Based Fault-Tolerant Binary Neural Networks (BNNs). A: 27th IEEE European Test Symposium (ETS). 2022,
Publisher versionhttps://ieeexplore.ieee.org/xpl/conhome/9810327/proceeding
Files | Description | Size | Format | View |
---|---|---|---|---|
PS1-4.pdf | 572,4Kb | Restricted access |