Understanding weight-magnitude hyperparameters in training binary networks

Understanding weight-magnitude hyperparameters in training binary networks

Quist, Joris (TU Delft Electrical Engineering, Mathematics and Computer Science)

Li, Y. (mentor)
van Gemert, J.C. (mentor)
Lofi, C. (graduation committee)

Delft University of Technology

Computer Science | Data Science and Technology

2022-11-07

Binary Neural Networks (BNNs) are compact and efficient by using binary weights instead of real-valued weights. Current BNNs use latent real-valued weights during training, where several training hyper-parameters are inherited from real-valued networks. The interpretation of several of these hyperparameters is based on the magnitude of the real-valued weights. For BNNs, however, the magnitude of binary weights is not meaningful, and thus it is unclear what these hyperparameters actually do. One example is weight-decay, which aims to keep the magnitude of real-valued weights small. Other examples are latent weight initialization, the learning rate, and learning rate decay, which influence the magnitude of the real-valued weights. The magnitude is interpretable for real-valued weights, but loses its meaning for binary weights.
In this paper we offer a new interpretation of these magnitude-based hyperparameters based on higher-order gradient filtering during network optimization. Our analysis makes it possible to understand how magnitude-based hyperparameters influence the training of binary networks which allows for new optimization filters specifically designed for binary neural networks that are independent of their real-valued interpretation. Moreover, our improved understanding reduces the number of hyperparameters, which in turn eases the hyperparameter tuning effort which may lead to better hyperparameter values for improved accuracy.

Deep Learning
Binary Neural Networks
Hyperparameter Optimization

http://resolver.tudelft.nl/uuid:fe011a49-787e-49cd-a5b3-2b63a8c8c4ef

Student theses

master thesis

© 2022 Joris Quist