The HSIC Bottleneck: Deep Learning without Back-Propagation
This work addresses the challenge of gradient issues in deep learning for researchers and practitioners, offering an alternative training method that is incremental but with potential practical benefits.
The authors tackled the problem of training deep neural networks without backpropagation by introducing the HSIC bottleneck, which mitigates exploding and vanishing gradients and achieves performance comparable to backpropagation on MNIST, FashionMNIST, and CIFAR10 classification tasks.
We introduce the HSIC (Hilbert-Schmidt independence criterion) bottleneck for training deep neural networks. The HSIC bottleneck is an alternative to the conventional cross-entropy loss and backpropagation that has a number of distinct advantages. It mitigates exploding and vanishing gradients, resulting in the ability to learn very deep networks without skip connections. There is no requirement for symmetric feedback or update locking. We find that the HSIC bottleneck provides performance on MNIST/FashionMNIST/CIFAR10 classification comparable to backpropagation with a cross-entropy target, even when the system is not encouraged to make the output resemble the classification labels. Appending a single layer trained with SGD (without backpropagation) to reformat the information further improves performance.