Theory

Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances ICML 2021 Global Optimality Beyond Two Layers: Training Deep ReLU Networks via Convex Programs ICML 2021 Guided Integrated Gradients: An Adaptive Path Method for Removing Noise CVPR 2021 Hessian Eigenspectra of More Realistic Nonlinear Models NeurIPS 2021 How Does Topology Influence Gradient Propagation and Model Performance of Deep Networks With DenseNet-Type Skip Connections? CVPR 2021 How Many Layers and Why? An Analysis of the Model Depth in Transformers ACL 2021 Hybrid Rotation Averaging: A Fast and Robust Rotation Averaging Approach CVPR 2021 Imitating Deep Learning Dynamics via Locally Elastic Stochastic Differential Equations NeurIPS 2021 Implicit Regularization in Tensor Factorization ICML 2021 Incremental Learning via Rate Reduction CVPR 2021 Interpreting Super-Resolution Networks With Local Attribution Maps CVPR 2021 Intrinsic Dimensionality Explains the Effectiveness of Language Model Fine-Tuning ACL 2021 Large Norms of CNN Layers Do Not Hurt Adversarial Robustness AAAI 2021 Last iterate convergence of SGD for Least-Squares in the Interpolation regime. NeurIPS 2021 Learning Curves for Analysis of Deep Networks ICML 2021 Learning Theory Can (Sometimes) Explain Generalisation in Graph Neural Networks NeurIPS 2021 Learning with Gradient Descent and Weakly Convex Losses AISTATS 2021 Linear Convergence of Gradient Methods for Estimating Structured Transition Matrices in High-dimensional Vector Autoregressive Models NeurIPS 2021 Local Signal Adaptivity: Provable Feature Learning in Neural Networks Beyond Kernels NeurIPS 2021 Modeling Trajectories with Neural Ordinary Differential Equations IJCAI 2021 Model, sample, and epoch-wise descents: exact solution of gradient flow in the random feature model NeurIPS 2021 Momentum Residual Neural Networks ICML 2021 Multiplicative Noise and Heavy Tails in Stochastic Optimization ICML 2021 Neural Splines: Fitting 3D Surfaces With Infinitely-Wide Neural Networks CVPR 2021 Noether’s Learning Dynamics: Role of Symmetry Breaking in Neural Networks NeurIPS 2021