Can I read BiPer: Binary Neural Networks using a Periodic Function on EtoBox?

BiPer: Binary Neural Networks using a Periodic Function by Vargas, Edwin; Correa, Claudia; Hinojosa, Carlos; Arguello, Henry is a scholarly article available to read on EtoBox.

What is BiPer: Binary Neural Networks using a Periodic Function about?

Quantized neural networks employ reduced precision representations for both weights and activations. This quantization process significantly reduces the memory requirements and computational complexity of the network. Binary Neural Networks (BNNs) are the extreme quantization case, representing values with just one bit. Since the sign function is typically used to map real values to binary values, smooth approximations are introduced to mimic the gradients during error backpropagation. Thus, the mismatch between the forward and backward models corrupts the direction of the gradient, causing training inconsistency problems and performance degradation. In contrast to current BNN approaches, we propose to employ a binary periodic (BiPer) function during binarization. Specifically, we use a square wave for the forward pass to obtain the binary values and employ the trigonometric sine function with the same period of the square wave as a differentiable surrogate during the backward pass. We demonstrate that this approach can control the quantization error by using the frequency of the periodic function and improves network performance. Extensive experiments validate the effectiveness of

Author

Vargas, Edwin; Correa, Claudia; Hinojosa, Carlos; Arguello, Henry

Published

2024

Language

EN