Data vs classifiers, who wins?

Abstract

The classification experiments covered by machine learning (ML) are composed by two important parts: the data and the algorithm. As they are a fundamental part of the problem, both must be considered when evaluating a model's performance against a benchmark. The best classifiers need robust benchmarks to be properly evaluated. For this, gold standard benchmarks such as OpenML-CC18 are used. However, data complexity is commonly not considered along with the model during a performance evaluation. Recent studies employ Item Response Theory (IRT) as a new approach to evaluating datasets and algorithms, capable of evaluating both simultaneously. This work presents a new evaluation methodology based on IRT and Glicko-2, jointly with the decodIRT tool developed to guide the estimation of IRT in ML. It explores the IRT as a tool to evaluate the OpenML-CC18 benchmark for its algorithmic evaluation capability and checks if there is a subset of datasets more efficient than the original benchmark. Several classifiers, from classics to ensemble, are also evaluated using the IRT models. The Glicko-2 rating system was applied together with IRT to summarize the innate ability and classifiers performance. It was noted that not all OpenML-CC18 datasets are really useful for evaluating algorithms, where only 10% were rated as being really difficult. Furthermore, it was verified the existence of a more efficient subset containing only 50% of the original size. While Randon Forest was singled out as the algorithm with the best innate ability.

HPOBench: A Collection of Reproducible Multi-Fidelity Benchmark Problems for HPO

“…We showed that advanced HPO methods are preferable over RS and HB baselines, and that multi-fidelity extensions of popular optimizers improve over their black-box version. Lastly, to reduce computational effort, we would like to study whether we can learn which benchmarks are hard and whether it suffices by executing only a representative subset of the benchmarks [16].…”