LightAutoML: AutoML Solution for a Large Financial Services Ecosystem

Vakhrushev, Anton; Ryzhkov, Alexander V.; Savchenko, Maxim; Simakov, D.; Damdinov, Rinchin; Tuzhilin, Alexander

doi:10.48550/arxiv.2109.01528

Cited by 10 publications

(12 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GAMA [41], and LightAutoML [42] will also be used to test their different approaches with our data sets.…”

Section: Automlmentioning

confidence: 99%

A practical evaluation of AutoML tools for binary, multiclass, and multilabel classification

Aragão¹,

Afonso²,

Ferraz³

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Automated Machine Learning (AutoML) is a recent technology that provides speed to machine learning iterations and allows individuals with less experience to take advantage of existing tools. Due to several frameworks with different features, deciding the best option to solve each classification problem becomes difficult. It is necessary to consider aspects such as performance metrics and time when choosing the algorithm to reduce the demand for highly technical, specific knowledge in the subject. There are some comparisons of AutoML tools and approaches that perform tests in the area of data preprocessing, model selection, and hyperparameter optimization. However, most of these studies focus on binary and multiclass classification, not covering multilabel classifications and consequently not exploiting the full potential of the tools. In this paper, a comparative study between multiple AutoML tools is performed related to the features, architecture, capabilities, and results achieved on binary, multiclass, and multilabel classification problems from experimentation on various data sets.</p>

show abstract

“…GAMA [41], and LightAutoML [42] will also be used to test their different approaches with our data sets.…”

Section: Automlmentioning

confidence: 99%

A practical evaluation of AutoML tools for binary, multiclass, and multilabel classification

Aragão¹,

Afonso²,

Ferraz³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The first frameworks that became well-known are H2O [15], TPOT [14] and Auto-sklearn [4]. As more novel AutoML solutions, Auto-Gluon [3] and LAMA [36] can be noted. Also, there are a lot of other AutoML tools with various specific features [17].…”

Section: Related Workmentioning

confidence: 99%

Improvement of Computational Performance of Evolutionary AutoML in a Heterogeneous Environment

Nikitin¹,

Teryoshkin²,

Pokrovskii³

et al. 2023

Preprint

View full text Add to dashboard Cite

Resource-intensive computations are a major factor that limits the effectiveness of automated machine learning solutions. In the paper, we propose a modular approach that can be used to increase the quality of evolutionary optimization for modelling pipelines with a graph-based structure. It consists of several stages -parallelization, caching and evaluation. Heterogeneous and remote resources can be involved in the evaluation stage. The conducted experiments confirm the correctness and effectiveness of the proposed approach. The implemented algorithms are available as a part of the open-source framework FEDOT.

show abstract

“…LIGHTAUTOML is specifically designed with applications in the financial services industry in mind (Vakhrushev et al, 2021). In this framework, pipelines are designed for quick inference and interpretability.…”

Section: Lightautomlmentioning

confidence: 99%

AMLB: an AutoML Benchmark

Gijsbers¹,

Bueno²,

Coors³

et al. 2022

Preprint

View full text Add to dashboard Cite

Comparing different AutoML frameworks is notoriously challenging and often done incorrectly. We introduce an open and extensible benchmark that follows best practices and avoids common mistakes when comparing AutoML frameworks. We conduct a thorough comparison of 9 well-known AutoML frameworks across 71 classification and 33 regression tasks. The differences between the AutoML frameworks are explored with a multi-faceted analysis, evaluating model accuracy, its trade-offs with inference time, and framework failures. We also use Bradley-Terry trees to discover subsets of tasks where the relative Au-toML framework rankings differ. The benchmark comes with an open-source tool that integrates with many AutoML frameworks and automates the empirical evaluation process end-to-end: from framework installation and resource allocation to in-depth evaluation. The benchmark uses public data sets, can be easily extended with other AutoML frameworks and tasks, and has a website with up-to-date results.

show abstract

LightAutoML: AutoML Solution for a Large Financial Services Ecosystem

Cited by 10 publications

References 18 publications

A practical evaluation of AutoML tools for binary, multiclass, and multilabel classification

A practical evaluation of AutoML tools for binary, multiclass, and multilabel classification

Improvement of Computational Performance of Evolutionary AutoML in a Heterogeneous Environment

AMLB: an AutoML Benchmark

Contact Info

Product

Resources

About