Adaptive Signal Variances: CNN Initialization Through Modern Architectures

Deep convolutional neural networks (CNN) have achieved the unwavering confidence in its performance on image processing tasks. The CNN architecture constitutes a variety of different types of layers including the convolution layer and the max-pooling layer. CNN practitioners widely understand the fact that the stability of learning depends on how to initialize the model parameters in each layer. Nowadays, no one doubts that the de facto standard scheme for initialization is the so-called Kaiming initialization that has been developed by He et al. The Kaiming scheme was derived from a much simpler model than the currently used CNN structure having evolved since the emergence of the Kaiming scheme. The Kaiming model consists only of the convolution and fully connected layers, ignoring the max-pooling layer and the global average pooling layer. In this study, we derived the initialization scheme again not from the simplified Kaiming model, but precisely from the modern CNN architectures, and empirically investigated how the new initialization method performs compared to the de facto standard ones that are widely used today.

Frank-Wolfe algorithm for learning SVM-type multi-category classifiers

Multi-category support vector machine (MC-SVM) is one of the most popular machine learning algorithms. There are lots of variants of MC-SVM, although different optimization algorithms were developed for different learning machines. In this study, we developed a new optimization algorithm that can be applied to many of MC-SVM variants. The algorithm is based on the Frank-Wolfe framework that requires two subproblems, direction finding and line search, in each iteration. The contribution of this study is the discovery that both subproblems have a closed form solution if the Frank-Wolfe framework is applied to the dual problem. Additionally, the closed form solutions on both for the direction finding and for the line search exist even for the Moreau envelopes of the loss functions. We use several large datasets to demonstrate that the proposed optimization algorithm converges rapidly and thereby improves the pattern recognition performance.