Significance: While breakthroughs in machine learning and artificial intelligence are changing society, our fundamental understanding has lagged behind. It is traditionally believed that fitting models to the training data exactly is to be avoided as it leads to poor performance on unseen data. However, powerful modern classifiers frequently have near-perfect fit in training, a disconnect that spurred recent intensive research and controversy on whether theory provides practical insights. In this work, we show how classical theory and modern practice can be reconciled within a single unified performance curve and propose a mechanism underlying its emergence. We believe this previously unknown pattern connecting the structure and performance of learning architectures will help shape design and understanding of learning algorithms.
In the past decade the mathematical theory of machine learning has lagged far behind the triumphs of deep neural networks on practical challenges. However, the gap between theory and practice is gradually starting to close. In this paper I will attempt to assemble some pieces of the remarkable and still incomplete mathematical mosaic emerging from the efforts to understand the foundations of deep learning. The two key themes will be interpolation and its sibling over-parametrization. Interpolation corresponds to fitting data, even noisy data, exactly. Over-parametrization enables interpolation and provides flexibility to select a suitable interpolating model.As we will see, just as a physical prism separates colours mixed within a ray of light, the figurative prism of interpolation helps to disentangle generalization and optimization properties within the complex picture of modern machine learning. This article is written in the belief and hope that clearer understanding of these issues will bring us a step closer towards a general theory of deep learning and machine learning.
Understanding how neural networks learn features, or relevant patterns in data, for prediction is necessary for their reliable use in technological and scientific applications. In this work, we presented a unifying mathematical mechanism, known as average gradient outer product (AGOP), that characterized feature learning in neural networks. We provided empirical evidence that AGOP captured features learned by various neural network architectures, including transformer-based language models, convolutional networks, multilayer perceptrons, and recurrent neural networks. Moreover, we demonstrated that AGOP, which is backpropagation-free, enabled feature learning in machine learning models, such as kernel machines, that a priori could not identify task-specific features. Overall, we established a fundamental mechanism that captured feature learning in neural networks and enabled feature learning in general machine learning models.
25 March 2019 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Radial Basis Function (RBF) networks are a classical family of algorithms for supervised learning. The most popular approach for training RBF networks has relied on kernel methods using regularization based on a norm in a Reproducing Kernel Hilbert Space (RKHS), which is a principled and empirically successful framework. In this paper we aim to revisit some of the older approaches to training the RBF networks from a more modern perspective. Specifically, we analyze two common regularization procedures, one based on the square norm of the coefficients in the network and another one using centers obtained by $k$ -means clustering. We show that both of these RBF methods can be recast as certain data-dependent kernels. We provide a theoretical analysis of these methods as well as a number of experimental results, pointing out very competitive experimental performance as well as certain advantages over the standard kernel methods in terms of both flexibility (incorporating of unlabeled data) and computational complexity. Finally, our results shed light on some impressive recent successes of using soft $k$ -means features for image recognition and other tasks.
23 May 2014 in Journal of Neurodevelopmental Disorders
It is commonly reported that children with autism spectrum disorder (ASD) exhibit hyper-reactivity or hypo-reactivity to sensory stimuli. Electroencephalography (EEG) is commonly used to study neural sensory reactivity, suggesting that statistical analysis of EEG recordings is a potential means of automatic classification of the disorder. EEG recordings taken from children, however, are frequently contaminated with large amounts of noise, making analysis difficult. In this paper, we present a method for the automatic extraction of noise-robust EEG features, which serve to quantify neural sensory reactivity. We show the efficacy of a system for the classification of ASD using these features.