Research map / Probabilistic and scientific modeling / Bayesian optimization
Scalable Hyperparameter Transfer Learning
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 |
|---|---|---|---|
| transfer from 29 related tasks (warm-start) instead of single-task (ablation) | ABLR NN · Parametrized quadratics | convergence speed | Helped |
| add task context vector as input (ablation) | multi-task ABLR · Parametrized quadratics | convergence speed | No clear effect |
| replace learned NN features (D=50) with random Fourier features RKS (D=100) (ablation) | single-task ABLR · Parametrized quadratics and OpenML | convergence speed | No clear effect |
| add auxiliary signals (train loss, cost, epochwise loss) as extra tasks (ablation) | ABLR NN · LIBSVM feedforward NN tuning | test error convergence | Helped |
| transfer learning with ABLR vs GP transfer | ABLR NN transfer · Parametrized quadratics | convergence speed | Mixed |
| multi-task ABLR transfer vs DNGO and BOHAMIANN transfer | ABLR NN transfer · Parametrized quadratics | transfer gain | Helped |
Closest papers by meaning
- Efficient Hyperparameter Optimization with Adaptive Fidelity Identification (CVPR 2024 · similarity 0.88)
- BOFormer: Learning to Solve Multi-Objective Bayesian Optimization via Non-Markovian RL (ICLR 2025 · similarity 0.87)
- Monte Carlo Tree Search based Space Transfer for Black Box Optimization (NeurIPS 2024 · similarity 0.87)
- Batched Energy-Entropy acquisition for Bayesian Optimization (NeurIPS 2024 · similarity 0.87)
- Posterior Inference with Diffusion Models for High-dimensional Black-box Optimization (ICML 2025 · similarity 0.87)
- Bayesian Optimisation with Unknown Hyperparameters: Regret Bounds Logarithmically Closer to Optimal (NeurIPS 2024 · similarity 0.87)
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.