Research map / Large language models / Agents, code and math
Pipeline Combinators for Gradual AutoML
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 Lale combinators instead of sklearn for T1 pipeline inspection | Lale · User study T1 | T1 correct: 89% vs 56% | Helped |
| use Lale combinators instead of sklearn for T4 search task | Lale · User study T4 | T4 correct: 100% vs 78% | Helped |
| use Lale or-combinator and AutoML search instead of sklearn | Lale · User study T4 | LOC median: 10 vs 23 | Helped |
| Lale instead of sklearn on total task time | Lale · User study | total time mean: 17:36 vs 19:26 | Helped |
| Lale instead of sklearn on refining and debugging pipelines | Lale · User study T2/T3 | task correctness | No clear effect |
| Lale-Auto instead of auto-sklearn, all 14 datasets | Lale-Auto · 14 OpenML datasets | average accuracy: 0.7% better | Helped |
| Lale-Auto instead of auto-sklearn, excluding shuttle | Lale-Auto · 13 OpenML datasets | average accuracy: 0.3% worse | No clear effect |
| Lale-Auto instead of auto-sklearn on shuttle | Lale-Auto · OpenML shuttle | test accuracy: 87.3 -> 100.0 | Helped |
| Lale-Auto instead of auto-sklearn on blood | Lale-Auto · OpenML blood | test accuracy: 77.9 -> 75.3 | Hurt |
| Lale-Auto instead of auto-sklearn on jungle-chess | Lale-Auto · OpenML jungle-chess | test accuracy: 88.1 -> 92.4 | Helped |
| Lale-Auto instead of auto-sklearn on car | Lale-Auto · OpenML car | test accuracy: 99.4 -> 97.7 | Hurt |
| Lale-Auto instead of auto-sklearn on australian | Lale-Auto · OpenML australian | test accuracy: 85.1 -> 86.2 | Helped |
Closest papers by meaning
- AutoMMLab: Automatically Generating Deployable Models from Language Instructions for Computer Vision Tasks (AAAI 2025 · similarity 0.85)
- AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML (ICML 2025 · similarity 0.85)
- ADELA: Accelerating Evolutionary Design of Machine Learning Pipelines with the Accompanying Surrogate Model (AAAI 2025 · similarity 0.85)
- Generalized Convergence Analysis of Tsetlin Automaton Based Algorithms: A Probabilistic Approach to Concept Learning (AAAI 2025 · similarity 0.83)
- AutoG: Towards automatic graph construction from tabular data (ICLR 2025 · similarity 0.82)
- LILO: Learning Interpretable Libraries by Compressing and Documenting Code (ICLR 2024 · similarity 0.82)
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.