2 papers
cs.CV2024
Efficiera Residual Networks: Hardware-Friendly Fully Binary Weight with 2-bit Activation Model Achieves Practical ImageNet Accuracy
Shuntaro Takahashi, Takuya Wakisaka, Hiroyuki Tokunaga
The edge-device environment imposes severe resource limitations, encompassing computation costs, hardware resource usage, and energy consumption for deploying deep neural network m…
cs.LG2024
Pixel Embedding: Fully Quantized Convolutional Neural Network with Differentiable Lookup Table
Hiroyuki Tokunaga, Joel Nicholls, Daria Vazhenina +1
By quantizing network weights and activations to low bitwidth, we can obtain hardware-friendly and energy-efficient networks. However, existing quantization techniques utilizing th…