Optimization

LocalNewton: Reducing communication rounds for distributed learning UAI 2021 Loss Surface Simplexes for Mode Connecting Volumes and Fast Ensembling ICML 2021 Low-Rank Sinkhorn Factorization ICML 2021 Low Resource Quadratic Forms for Knowledge Graph Embeddings EMNLP 2021 LQF: Linear Quadratic Fine-Tuning CVPR 2021 Marginalized Stochastic Natural Gradients for Black-Box Variational Inference ICML 2021 MARINA: Faster Non-Convex Distributed Learning with Compression ICML 2021 Meta-Cal: Well-controlled Post-hoc Calibration by Ranking ICML 2021 MetricOpt: Learning To Optimize Black-Box Evaluation Metrics CVPR 2021 M-FAC: Efficient Matrix-Free Approximations of Second-Order Information NeurIPS 2021 Minibatch and Momentum Model-based Methods for Stochastic Weakly Convex Optimization NeurIPS 2021 MongeNet: Efficient Sampler for Geometric Deep Learning CVPR 2021 Monte Carlo Scene Search for 3D Scene Understanding CVPR 2021 Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices NeurIPS 2021 MTAdam: Automatic Balancing of Multiple Training Loss Terms EMNLP 2021 Multi-Layer Random Perturbation Training for improving Model Generalization Efficiently EMNLP 2021 Multi-Level Local SGD: Distributed SGD for Heterogeneous Hierarchical Networks ICLR 2021 NetAdaptV2: Efficient Neural Architecture Search With Fast Super-Network Training and Architecture Optimization CVPR 2021 Neural Architecture Search as Sparse Supernet AAAI 2021 Neural Reprojection Error: Merging Feature Learning and Camera Pose Estimation CVPR 2021 Newton Method over Networks is Fast up to the Statistical Precision ICML 2021 Non-asymptotic Convergence of Adam-type Reinforcement Learning Algorithms under Markovian Sampling AAAI 2021 Nonsmooth Implicit Differentiation for Machine-Learning and Optimization NeurIPS 2021 NUQSGD: Provably Communication-efficient Data-parallel SGD via Nonuniform Quantization JMLR 2021 On Orthogonality Constraints for Transformers ACL 2021