Neural Network Optimization

A Variational Perspective on High-Resolution ODEs NeurIPS 2023 B2T Connection: Serving Stability and Performance in Deep Transformers ACL 2023 Backforward Propagation (Student Abstract) AAAI 2023 Backpropagation-Free Deep Learning with Recursive Local Representation Alignment AAAI 2023 Backpropagation of Unrolled Solvers with Folded Optimization IJCAI 2023 Backpropagation through Combinatorial Algorithms: Identity with Projection Works ICLR 2023 BADGE: Speeding Up BERT Inference after Deployment via Block-wise Bypasses and Divergence-based Early Exiting ACL 2023 BAT: Boundary aware transducer for memory-efficient and low-latency ASR Interspeech 2023 Behind the Scenes of Gradient Descent: A Trajectory Analysis via Basis Function Decomposition ICLR 2023 Benign Overfitting in Deep Neural Networks under Lazy Training ICML 2023 Better Loss Landscape Visualization for Deep Neural Networks with Trajectory Information ACML 2023 Beyond Lipschitz Smoothness: A Tighter Analysis for Nonconvex Optimization ICML 2023 Beyond NTK with Vanilla Gradient Descent: A Mean-Field Analysis of Neural Networks with Polynomial Width, Samples, and Time NeurIPS 2023 Beyond the Edge of Stability via Two-step Gradient Updates ICML 2023 Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels ICML 2023 Beyond Uniform Lipschitz Condition in Differentially Private Optimization ICML 2023 Beyond Uniform Smoothness: A Stopped Analysis of Adaptive SGD COLT 2023 Bidirectional Looking with A Novel Double Exponential Moving Average to Adaptive and Non-adaptive Momentum Optimizers ICML 2023 Bifurcations and loss jumps in RNN training NeurIPS 2023 Binarizing Sparse Convolutional Networks for Efficient Point Cloud Analysis CVPR 2023 Birder: Communication-Efficient 1-bit Adaptive Optimizer for Practical Distributed DNN Training NeurIPS 2023 BiSLS/SPS: Auto-tune Step Sizes for Stable Bi-level Optimization NeurIPS 2023 Block Low-Rank Preconditioner with Shared Basis for Stochastic Optimization NeurIPS 2023 Boosting Adversarial Transferability by Achieving Flat Local Maxima NeurIPS 2023 Boosting Learning for LDPC Codes to Improve the Error-Floor Performance NeurIPS 2023