Research map / Optimization / Nonconvex and smooth optimization
Simple Stochastic Gradient Methods for Non-Smooth Non-Convex Regularized Optimization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Mini-batch stochastic gradient algorithm (MBSGA) with Moreau envelope based smooth majorant instead of SSDC-SPG | non-smooth non-convex regularized stochastic optimization · theoretical gradient call complexity | gradient call complexity to reach eps-stationary point: O(eps^-5) vs O(eps^-8) | Helped |
| Variance reduced SGD (VRSGA) instead of SSDC-SVRG for finite-sum problems | non-smooth non-convex regularized finite-sum optimization · theoretical gradient call complexity | gradient call complexity: O(n^2/3 eps^-3) vs O(n eps^-4) | Helped |
| Simplest algorithm (MBSGA) compared with SSDC-SPG/SSDC-SVRG on binary classification with Lorenz loss and log-sum regularizer | MBSGA · a9a and MNIST objective log(h(w)) vs time | log objective vs wall-clock time | Helped |
| Variance reduction (VRSGA) added on top of the finite-sum method compared with MBSGA | VRSGA · a9a and MNIST objective log(h(w)) vs time | log objective vs wall-clock time | No clear effect |
Closest papers by meaning
- Non-asymptotic Analysis of Stochastic Methods for Non-Smooth Non-Convex Regularized Problems (NeurIPS 2019 · similarity 0.93)
- Stochastic Optimization for DC Functions and Non-smooth Non-convex Regularizers with Non-asymptotic Convergence (ICML 2019 · similarity 0.92)
- Proximal Stochastic Methods for Nonsmooth Nonconvex Finite-Sum Optimization (NeurIPS 2016 · similarity 0.92)
- Universal Gradient Methods for Stochastic Convex Optimization (ICML 2024 · similarity 0.91)
- Zeroth-order (Non)-Convex Stochastic Optimization via Conditional Gradient and Gradient Updates (NeurIPS 2018 · similarity 0.91)
- Decentralized Gradient-Free Methods for Stochastic Non-smooth Non-convex Optimization (AAAI 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.