16-bit posits train MNIST and Fashion-MNIST feedforward networks with less accuracy loss than 16-bit floating point, and 5 to 8 bit posits yield better inference accuracy and energy-delay tradeoffs than float or fixed point.
Imagenet classification with deep convolutional neural networks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Cheetah: Mixed Low-Precision Hardware & Software Co-Design Framework for DNNs on the Edge
16-bit posits train MNIST and Fashion-MNIST feedforward networks with less accuracy loss than 16-bit floating point, and 5 to 8 bit posits yield better inference accuracy and energy-delay tradeoffs than float or fixed point.