Authors apply a consistent methodology to benchmark physical performance metrics across neural network architectures and device technologies, identifying promising combinations.
They are based on traditional digital chips and in this sense are different from other neuromorphic hardware
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.ET 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Benchmarking Physical Performance of Neural Inference Circuits
Authors apply a consistent methodology to benchmark physical performance metrics across neural network architectures and device technologies, identifying promising combinations.