AlphaD3M: Machine Learning Pipeline Synthesis
This addresses the efficiency and explainability challenges in AutoML for researchers and practitioners, though it appears incremental as it builds on existing AutoML methods with improvements in speed and transparency.
The paper tackles the problem of automating machine learning pipeline design by introducing AlphaD3M, an AutoML system that achieves competitive performance compared to state-of-the-art systems like Autosklearn, Autostacker, and TPOT, while being an order of magnitude faster, reducing computation time from hours to minutes.
We introduce AlphaD3M, an automatic machine learning (AutoML) system based on meta reinforcement learning using sequence models with self play. AlphaD3M is based on edit operations performed over machine learning pipeline primitives providing explainability. We compare AlphaD3M with state-of-the-art AutoML systems: Autosklearn, Autostacker, and TPOT, on OpenML datasets. AlphaD3M achieves competitive performance while being an order of magnitude faster, reducing computation time from hours to minutes, and is explainable by design.