JMLR 2021 papers on Optimization
43 of the 285 papers published in JMLR 2021 (15.1%) are on Optimization, compared with 10.2% at JMLR 2020. They fall into 4 topics.
See JMLR 2021 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Nonconvex and smooth optimization · 24 papers
- A Lyapunov Analysis of Accelerated Methods in Optimization (44 citations)
- Consensus-Based Optimization on the Sphere: Convergence to Global Minimizers and Machine Learning (38 citations)
- A Unified Convergence Analysis for Shuffling-Type Gradient Methods (34 citations)
- An Inertial Newton Algorithm for Deep Learning (34 citations)
- Online stochastic gradient descent on non-convex losses from high-dimensional inference (27 citations)
- Optimization with Momentum: Dynamical, Control-Theoretic, and Symplectic Perspectives (27 citations)
- Stochastic Proximal AUC Maximization (18 citations)
- L-SVRG and L-Katyusha with Arbitrary Sampling (17 citations)
- Generalization Performance of Multi-pass Stochastic Gradient Descent with Convex Loss Functions (16 citations)
- A Unified Analysis of First-Order Methods for Smooth Games via Integral Quadratic Constraints (15 citations)
- Replica Exchange for Non-Convex Optimization (15 citations)
- From Low Probability to High Confidence in Stochastic Convex Optimization (14 citations)
and 12 more in the interactive map.
Implicit bias and generalization · 9 papers
- Accelerating Ill-Conditioned Low-Rank Matrix Estimation via Scaled Gradient Descent (63 citations)
- Finite-sample Analysis of Interpolating Linear Classifiers in the Overparameterized Regime (59 citations)
- Convex Geometry and Duality of Over-parameterized Neural Networks (25 citations)
- Is SGD a Bayesian sampler? Well, almost (24 citations)
- On ADMM in Deep Learning: Convergence and Saturation-Avoidance (22 citations)
- Continuous Time Analysis of Momentum Methods (20 citations)
- Gradient Methods Never Overfit On Separable Data (20 citations)
- When Does Gradient Descent with Logistic Loss Find Interpolating Two-Layer Networks? (8 citations)
- An algorithmic view of L2 regularization and some path-following algorithms
Distributed and compressed · 5 papers
- Cooperative SGD: A Unified Framework for the Design and Analysis of Local-Update SGD Algorithms (130 citations)
- NUQSGD: Provably Communication-efficient Data-parallel SGD via Nonuniform Quantization (30 citations)
- COKE: Communication-Censored Decentralized Kernel Learning (24 citations)
- Variance Reduced Median-of-Means Estimator for Byzantine-Robust Distributed Inference (14 citations)
- DeEPCA: Decentralized Exact PCA with Linear Convergence Rate (10 citations)
Optimizers · 5 papers
- The ensmallen library for flexible numerical optimization (10 citations)
- Langevin Monte Carlo: random coordinate descent and variance reduction (1 citation)
- Black-Box Reductions for Zeroth-Order Gradient Algorithms to Achieve Lower Query Complexity
- Multilevel Monte Carlo Variational Inference
- Quasi-Monte Carlo Quasi-Newton in Variational Bayes
