A tiny CNN front-end plus an RRAM-CMOS analog content-addressable memory back-end classifies greyscale CIFAR-10 at 70.9% accuracy while claiming roughly 800x lower inference energy than a ResNet-50 teacher.
Hardware-aware approach to deep neural network optimization,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
A Hybrid Edge Classifier: Combining TinyML-Optimised CNN with RRAM-CMOS ACAM for Energy-Efficient Inference
A tiny CNN front-end plus an RRAM-CMOS analog content-addressable memory back-end classifies greyscale CIFAR-10 at 70.9% accuracy while claiming roughly 800x lower inference energy than a ResNet-50 teacher.