Research map / Optimization / Nonconvex and smooth optimization
On Markov Chain Gradient Descent
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 |
|---|---|---|---|
| Use one continuing Markov-chain trajectory instead of independent fresh trajectories for each gradient | SGD on autoregressive-process recovery · Convex logistic and nonconvex squared logistic losses | total samples used | Helped |
| Use a non-reversible chain with added directed cycles instead of the reversible chain | MCGD on graph least squares · Least squares on graph with 20 nodes | convergence: 0.75 -> 0.66 | Helped |
| Use fresh trajectories with T=1 instead of a continuing Markov-chain trajectory | SGD on nonconvex recovery · Autoregressive-process recovery, nonconvex squared logistic loss | accuracy | Hurt |
| Use fresh trajectories with T=2 instead of a continuing Markov-chain trajectory | SGD on nonconvex recovery · Autoregressive-process recovery, nonconvex squared logistic loss | accuracy | Hurt |
| Use fresh trajectories with T=4 instead of a continuing Markov-chain trajectory | SGD on nonconvex recovery · Autoregressive-process recovery, nonconvex squared logistic loss | accuracy | Hurt |
Closest papers by meaning
- Stochastic Gradient Descent under Markovian Sampling Schemes (ICML 2023 · similarity 0.93)
- High-probability Convergence of Gradient Methods under Markovian Stochasticity (NeurIPS 2026 · similarity 0.92)
- Adapting to Mixing Time in Stochastic Optimization with Markovian Data (ICML 2022 · similarity 0.91)
- Improved Lower Bounds for First-order Stochastic Non-convex Optimization under Markov Sampling (ICML 2025 · similarity 0.89)
- Nonasymptotic Analysis of Stochastic Gradient Descent with the Richardson–Romberg Extrapolation (ICLR 2025 · similarity 0.88)
- The Global Convergence Time of Stochastic Gradient Descent in Non-Convex Landscapes: Sharp Estimates via Large Deviations (ICML 2025 · similarity 0.88)
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.