Theory

Reconciling Modern Deep Learning with Traditional Optimization Analyses: The Intrinsic Learning Rate NeurIPS 2020 Rethinking Bias-Variance Trade-off for Generalization of Neural Networks ICML 2020 Reverse Engineering Configurations of Neural Text Generation Models ACL 2020 RNNs can generate bounded hierarchical languages with optimal memory EMNLP 2020 Robust Design of Deep Neural Networks Against Adversarial Attacks Based on Lyapunov Theory CVPR 2020 Robustness Analysis of Non-Convex Stochastic Gradient Descent using Biased Expectations NeurIPS 2020 Sample Complexity Bounds for RNNs with Application to Combinatorial Graph Problems (Student Abstract) AAAI 2020 Sharp Representation Theorems for ReLU Networks with Precise Dependence on Depth NeurIPS 2020 Spectra of the Conjugate Kernel and Neural Tangent Kernel for linear-width neural networks NeurIPS 2020 Stable behaviour of infinitely wide deep neural networks AISTATS 2020 Stolen Probability: A Structural Weakness of Neural Language Models ACL 2020 TBT: Targeted Neural Network Attack With Bit Trojan CVPR 2020 TESA: Tensor Element Self-Attention via Matricization CVPR 2020 The asymptotic spectrum of the Hessian of DNN throughout training ICLR 2020 The HSIC Bottleneck: Deep Learning without Back-Propagation AAAI 2020 The Implicit and Explicit Regularization Effects of Dropout ICML 2020 The phase diagram of approximation rates for deep neural networks NeurIPS 2020 The Statistical Complexity of Early-Stopped Mirror Descent NeurIPS 2020 The Surprising Simplicity of the Early-Time Learning Dynamics of Neural Networks NeurIPS 2020 Transformation of ReLU-based recurrent neural networks from discrete-time to continuous-time ICML 2020 Understanding Architectures Learnt by Cell-based Neural Architecture Search ICLR 2020 Understanding Double Descent Requires A Fine-Grained Bias-Variance Decomposition NeurIPS 2020 Understanding Generalization in Deep Learning via Tensor Methods AISTATS 2020 Understanding the Difficulty of Training Transformers EMNLP 2020 Unique Properties of Flat Minima in Deep Networks ICML 2020