Research map / Optimization / Nonconvex and smooth optimization
Integration Methods and Optimization Algorithms
Damien Scieur, Vincent Roulet, Francis Bach, Alexandre d'Aspremont
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| replace Euler (gradient descent) with the two-step scheme M1 equal to Nesterov's method, using a larger integration step size | gradient flow discretization on L-smooth mu-strongly convex functions · theoretical convergence rate | convergence rate | Improved |
| tune two-step method parameters to the optimal step size (M2) instead of h-hat = 1/L (M1) | two-step linear multistep method on quadratics · linear ODE (quadratic minimization) | largest root modulus (rate) | Improved |
| use heavy ball (M2 = Polyak) on general smooth strongly convex functions rather than only quadratics | Polyak heavy ball method · general smooth strongly convex functions | convergence | Hurt |
| use optimal versus small step size in integration of a linear ODE | Euler, Nesterov and Polyak schemes · Figure 1 simulation | trajectory tracking of exact solution | Mixed |
Closest papers by meaning
- A Geometric Structure of Acceleration and Its Role in Making Gradients Small Fast (NeurIPS 2021 · similarity 0.88)
- Accelerating Optimization via Differentiable Stopping Time (NeurIPS 2025 · similarity 0.86)
- Optimal Acceleration for Minimax and Fixed-Point Problems is Not Unique (ICML 2024 · similarity 0.85)
- Optimizing $(L_0, L_1)$-Smooth Functions by Gradient Methods (ICLR 2025 · similarity 0.84)
- Nonlinearly Preconditioned Gradient Methods: Momentum and Stochastic Analysis (NeurIPS 2025 · similarity 0.84)
- Hamiltonian Descent Algorithms for Optimization: Accelerated Rates via Randomized Integration Time (NeurIPS 2025 · similarity 0.84)
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.
