FLAML: A Fast and Lightweight AutoML Library

By Chi Wang, Qingyun Wu, Markus Weimer, Erkang Zhu. Conference on Machine Learning and Systems, 2019

303 citations recorded by the source corpus.

We study the problem of using low computational cost to automate the choices of learners and hyperparameters for an ad-hoc training dataset and error metric, by conducting trials of different configurations on the given training data. We investigate the joint impact of multiple factors on both trial cost and model error, and propose several design guidelines. Following them, we build a fast and lightweight library FLAML which optimizes for low computational resource in finding accurate models. FLAML integrates several simple but effective search strategies into an adaptive system. It significantly outperforms top-ranked AutoML libraries on a large open source AutoML benchmark under equal, or sometimes orders of magnitude smaller budget constraints.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/10a0ea3e1b3202d1c28903cc823d5ab8e9d5780d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

OpenML Benchmarking Suites (2017, methodology): “The original AutoML benchmark suite has been used in multiple AutoML publications, either as is [LeDell and Poirier, 2020, Wang et al., 2021, Feurer et al., 2021a] or with modiﬁcations.”
DISTILL: Low-Overhead Data-Driven Techniques for Filtering and Costing Indexes for Scalable Index Tuning (2022, methodology): “We considered the following regression models: Logistic regression (LR), three tree-based ensemble models: Light GBM (LGBM), XGBoost, and Random Forest (RF) with hyperparameters tuned using FLAML [36].”
AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation (2023, unknown): “tune’ (Wang et al., 2021) can be used together with AutoGen to automate such tuning.”