P4ML: A Phased Performance-Based Pipeline Planner for Automated Machine Learning

How researchers see this paper

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

Benchmark and Survey of Automated Machine Learning Frameworks (2019, methodology): “These additional requirements can be detected by analyzing a candidate pipeline and matching the prerequisites of every stage with metafeatures of each feature in the data set (Gil et al., 2018; Nguyen et al., 2020).”
Survey on Automated Machine Learning (2019, background): “These additional requirements can be detected by analyzing a candidate pipeline and matching the prerequisites of every stage with meta-features of each feature in the data set (Gil et al., 2018).”
Meta-Learning: A Survey (2018, neutral): “, 2018), and can be leveraged in larger pipeline construction approaches, such as planning (Nguyen et al., 2014; Kietz et al., 2012; Gil et al., 2018; Wever et al., 2018) or evolutionary techniques (Olson et al.”
OpenML Benchmarking Suites (2017, background): “For example, Drori et al. [2018], Rakotoarison et al. [2019] and Gil et al. [2018], all published at the same workshop, each used different selections of datasets.”
Towards human-guided machine learning (2019, methodology): “Our AutoML system uses a phased approach to AutoML that can ingest naturally occurring data of any size and type, and use third-party data processing and modeling primitives [10].”
AVATAR - Machine Learning Pipeline Evaluation Using Surrogate Model (2020, methodology): “The final output of P4ML is a pipeline which is constructed by making an ensemble of other pipelines.”