Theory

Towards Better Understanding of Training Certifiably Robust Models against Adversarial Examples NeurIPS 2021 Towards Lower Bounds on the Depth of ReLU Neural Networks NeurIPS 2021 Towards Scalable Complete Verification of Relu Neural Networks via Dependency-based Branching IJCAI 2021 Towards Sharper Generalization Bounds for Structured Prediction NeurIPS 2021 Towards the evaluation of automatic simultaneous speech translation from a communicative perspective ACL 2021 Towards Tight Bounds on the Sample Complexity of Average-reward MDPs ICML 2021 Training deep residual networks for uniform approximation guarantees L4DC 2021 Training Neural Networks is ER-complete NeurIPS 2021 Train simultaneously, generalize better: Stability of gradient-based minimax learners ICML 2021 Transformer Interpretability Beyond Attention Visualization CVPR 2021 Trembling-Hand Perfection and Correlation in Sequential Games AAAI 2021 Turing Completeness of Bounded-Precision Recurrent Neural Networks NeurIPS 2021 Two Sides of Meta-Learning Evaluation: In vs. Out of Distribution NeurIPS 2021 Types of Out-of-Distribution Texts and How to Detect Them EMNLP 2021 Uncertainty Principles of Encoding GANs ICML 2021 Understanding End-to-End Model-Based Reinforcement Learning Methods as Implicit Parameterization NeurIPS 2021 Understanding Failures of Deep Networks via Robust Feature Extraction CVPR 2021 Understanding the Behaviour of Contrastive Loss CVPR 2021 Understanding the Dynamics of Gradient Flow in Overparameterized Linear models ICML 2021 Understanding the Limits of Unsupervised Domain Adaptation via Data Poisoning NeurIPS 2021 Understanding the Under-Coverage Bias in Uncertainty Estimation NeurIPS 2021 UnICORNN: A recurrent model for learning very long time dependencies ICML 2021 Uniform Consistency of Cross-Validation Estimators for High-Dimensional Ridge Regression AISTATS 2021 Uniform Convergence, Adversarial Spheres and a Simple Remedy ICML 2021 Uniform Convergence of Interpolators: Gaussian Width, Norm Bounds and Benign Overfitting NeurIPS 2021