HadaNets: Flexible Quantization Strategies for Neural Networks
This addresses energy efficiency for on-board processing in UAVs, offering incremental improvements in quantization and compression.
The paper tackles the problem of high energy consumption from memory accesses in deep neural networks for UAVs by introducing HadaNets, a flexible quantization method that pairs full precision and binary tensors, achieving a 7.43x model size reduction in ResNet-18 and a 10-fold performance increase in matrix multiplication.
On-board processing elements on UAVs are currently inadequate for training and inference of Deep Neural Networks. This is largely due to the energy consumption of memory accesses in such a network. HadaNets introduce a flexible train-from-scratch tensor quantization scheme by pairing a full precision tensor to a binary tensor in the form of a Hadamard product. Unlike wider reduced precision neural network models, we preserve the train-time parameter count, thus out-performing XNOR-Nets without a train-time memory penalty. Such training routines could see great utility in semi-supervised online learning tasks. Our method also offers advantages in model compression, as we reduce the model size of ResNet-18 by 7.43 times with respect to a full precision model without utilizing any other compression techniques. We also demonstrate a 'Hadamard Binary Matrix Multiply' kernel, which delivers a 10-fold increase in performance over full precision matrix multiplication with a similarly optimized kernel.