# Mining Robust Default Configurations for Resource-constrained AutoML

arXiv: https://arxiv.org/abs/2202.09927 · HTML: https://mlsanity.com/paper/2202.09927/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 26 references that are in the map.
- First posted on arXiv 20 Feb 2022.

## Builds on (most cited first, 9 of 9)

- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data](https://mlsanity.com/paper/2003.06505/) (2020)
- [FLAML: A Fast and Lightweight AutoML Library](https://mlsanity.com/paper/1911.04706/) (2019)
- [Automating biomedical data science through tree-based pipeline optimization](https://mlsanity.com/paper/1601.07925/) (2016)
- [Amazon SageMaker Autopilot: a white box AutoML solution at scale](https://mlsanity.com/paper/2012.08483/) (2020)
- [Cloudy with high chance of DBMS: A 10-year prediction for Enterprise-Grade ML](https://mlsanity.com/paper/1909.00084/) (2019)
- [Meta-Learning for Symbolic Hyperparameter Defaults](https://mlsanity.com/paper/2106.05767/) (2021)
- [Privileged Zero-Shot AutoML](https://mlsanity.com/paper/2106.13743/) (2021)
- [ChaCha for Online AutoML](https://mlsanity.com/paper/2106.04815/) (2021)

## Built on by (most cited first, 2 of 2)

- [Edge-Cloud Collaborative Computing on Distributed Intelligence and Model Optimization: A Survey](https://mlsanity.com/paper/2505.01821/) (2025)
- [How Many Validation Labels Do You Need? Exploring the Design Space of Label-Efficient Model Ranking](https://mlsanity.com/paper/2312.01619/) (2023)

## Abstract (arXiv)

Automatic machine learning (AutoML) is a key enabler of the mass deployment of the next generation of machine learning systems. A key desideratum for future ML systems is the automatic selection of models and hyperparameters. We present a novel method of selecting performant configurations for a given task by performing offline autoML and mining over a diverse set of tasks. By mining the training tasks, we can select a compact portfolio of configurations that perform well over a wide variety of tasks, as well as learn a strategy to select portfolio configurations for yet-unseen tasks. The algorithm runs in a zero-shot manner, that is without training any models online except the chosen one. In a compute- or time-constrained setting, this virtually instant selection is highly performant. Further, we show that our approach is effective for warm-starting existing autoML platforms. In both settings, we demonstrate an improvement on the state-of-the-art by testing over 62 classification and regression datasets. We also demonstrate the utility of recommending data-dependent default configurations that outperform widely used hand-crafted defaults.