Research map / Optimization / Nonconvex and smooth optimization
Hamiltonian descent for composite objectives
Brendan O'Donoghue, Chris J. Maddison
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 |
|---|---|---|---|
| worsen data conditioning from j=0 to j=20 (condition number 4.0 x 10^3 to 2.2 x 10^14) | Gradient descent and other first-order baselines · Regularized least-squares | convergence speed of primal objective error: 4.0 × 103 -> 2.2 × 1014 | Hurt |
| worsen data conditioning from j=0 to j=20 | Hamiltonian descent · Regularized least-squares | convergence of primal objective error and duality gap | No clear effect |
| use Hamiltonian descent instead of proximal gradient descent | Hamiltonian descent · Logistic regression with elastic net, condition number 1.0 × 108 | primal objective error | Improved |
Closest papers by meaning
- Time-Reversed Dissipation Induces Duality Between Minimizing Gradient Norm and Function Value (NeurIPS 2023 · similarity 0.88)
- Hamiltonian Descent Algorithms for Optimization: Accelerated Rates via Randomized Integration Time (NeurIPS 2025 · similarity 0.85)
- The Physical Systems Behind Optimization Algorithms (NeurIPS 2018 · similarity 0.85)
- Discrete Variational Calculus for Accelerated Optimization (JMLR 2023 · similarity 0.85)
- Two-timescale Extragradient for Finding Local Minimax Points (ICLR 2024 · similarity 0.84)
- Sharper Convergence Rates for Nonconvex Optimisation via Reduction Mappings (NeurIPS 2025 · similarity 0.83)
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.
