AutoML in Heavily Constrained Applications

TL;DR

This paper introduces CAML, a meta-learning based AutoML system that automatically adapts its configuration to specific tasks and constraints, improving pipeline effectiveness and efficiency.

Contribution

CAML is the first AutoML system that dynamically adjusts its parameters and incorporates user-defined constraints using meta-learning.

Findings

CAML effectively adapts its search strategy to different tasks.

It produces pipelines that satisfy user constraints with high predictive performance.

CAML outperforms static AutoML configurations in constrained environments.

Abstract

Optimizing a machine learning pipeline for a task at hand requires careful configuration of various hyperparameters, typically supported by an AutoML system that optimizes the hyperparameters for the given training dataset. Yet, depending on the AutoML system's own second-order meta-configuration, the performance of the AutoML process can vary significantly. Current AutoML systems cannot automatically adapt their own configuration to a specific use case. Further, they cannot compile user-defined application constraints on the effectiveness and efficiency of the pipeline and its generation. In this paper, we propose CAML, which uses meta-learning to automatically adapt its own AutoML parameters, such as the search strategy, the validation strategy, and the search space, for a task at hand. The dynamic AutoML strategy of CAML takes user-defined constraints into account and obtains constraint-satisfying pipelines with high predictive performance.