Theory

Spectrally-normalized margin bounds for neural networks NeurIPS 2017 SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability NeurIPS 2017 The Expressive Power of Neural Networks: A View from the Width NeurIPS 2017 The Marginal Value of Adaptive Gradient Methods in Machine Learning NeurIPS 2017 The Numerics of GANs NeurIPS 2017 Towards Generalization and Simplicity in Continuous Control NeurIPS 2017 Train longer, generalize better: closing the generalization gap in large batch training of neural networks NeurIPS 2017 Convolutional Rectifier Networks as Generalized Tensor Decompositions ICML 2016 Deep Learning Games NeurIPS 2016 Discrete Deep Feature Extraction: A Theory and New Architectures ICML 2016 Exponential expressivity in deep neural networks through transient chaos NeurIPS 2016 Globally Optimal Training of Generalized Polynomial Neural Networks with Nonlinear Spectral Methods NeurIPS 2016 On the Quality of the Initial Basin in Overspecified Neural Networks ICML 2016 On the Role of Nonlinear Transformations in Deep Neural Network Acoustic Models Interspeech 2016 Optimal Architectures in a Solvable Model of Deep Networks NeurIPS 2016 Orthogonal Random Features NeurIPS 2016 Residual Networks Behave Like Ensembles of Relatively Shallow Networks NeurIPS 2016 Toward Deeper Understanding of Neural Networks: The Power of Initialization and a Dual View on Expressivity NeurIPS 2016 A Stable Multi-Scale Kernel for Topological Machine Learning CVPR 2015 Understanding Image Representations by Measuring Their Equivariance and Equivalence CVPR 2015 Altitude Training: Strong Bounds for Single-Layer Dropout NeurIPS 2014 Identifying and attacking the saddle point problem in high-dimensional non-convex optimization NeurIPS 2014 On the Computational Efficiency of Training Neural Networks NeurIPS 2014 On the Number of Linear Regions of Deep Neural Networks NeurIPS 2014 Signal recovery from Pooling Representations ICML 2014