Filters








57 Hits in 4.0 sec

Automatically Debugging AutoML Pipelines Using Maro: ML Automated Remediation Oracle (Extended Version) [article]

Julian Dolby, Jason Tsay, Martin Hirzel
2022 arXiv   pre-print
We implemented our approach, which builds on a combination of AutoML and SMT, in a tool called Maro.  ...  Machine learning in practice often involves complex pipelines for data cleansing, feature engineering, preprocessing, and prediction.  ...  This paper presents a tool named Maro (ML Automated Remediation Oracle) that automatically debugs ML pipelines and generates remediated pipelines based on AutoML experiment results.  ... 
arXiv:2205.01311v1 fatcat:xvlk2rmtlnbvbo2gggvxz7tbzy

FLAML: A Fast and Lightweight AutoML Library [article]

Chi Wang, Qingyun Wu, Markus Weimer, Erkang Zhu
2021 arXiv   pre-print
Following them, we build a fast and lightweight library FLAML which optimizes for low computational resource in finding accurate models.  ...  It significantly outperforms top-ranked AutoML libraries on a large open source AutoML benchmark under equal, or sometimes orders of magnitude smaller budget constraints.  ...  ACKNOWLEDGMENTS The authors appreciate suggestions from Surajit Chaudhuri, Nadiia Chepurko, Alex Deng, Anshuman Dutt, Johannes Gehrke, Silu Huang, Christian Konig, and Haozhe Zhang.  ... 
arXiv:1911.04706v3 fatcat:uzkwcroe4rhzjdbgvb2mrzwnxa

Mining Robust Default Configurations for Resource-constrained AutoML [article]

Moe Kayali, Chi Wang
2022 arXiv   pre-print
We present a novel method of selecting performant configurations for a given task by performing offline autoML and mining over a diverse set of tasks.  ...  A key desideratum for future ML systems is the automatic selection of models and hyperparameters.  ...  We are also grateful to DevDiv and ML.NET teams for their feedback on the key points of this work.  ... 
arXiv:2202.09927v1 fatcat:46jdrg6xanefddr2rboutdaoaq

Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learning [article]

Matthias Feurer, Katharina Eggensperger, Stefan Falkner, Marius Lindauer, Frank Hutter
2021 arXiv   pre-print
Automated Machine Learning (AutoML) supports practitioners and researchers with the tedious task of designing machine learning pipelines and has recently achieved substantial success.  ...  Therefore, we also go one step further and study the design space of AutoML itself, proposing a solution towards truly hands-free AutoML.  ...  We furthermore thank all contributors to Auto-sklearn for their help in making it a useful AutoML tool and also thank Francisco Rivera for providing a Singularity integration for the AutoML benchmark.  ... 
arXiv:2007.04074v2 fatcat:qcrxuttjjfgnpnklql4ars4hbu

VolcanoML: Speeding up End-to-End AutoML via Scalable Search Space Decomposition [article]

Yang Li, Yu Shen, Wentao Zhang, Jiawei Jiang, Bolin Ding, Yaliang Li, Jingren Zhou, Zhi Yang, Wentao Wu, Ce Zhang, Bin Cui
2021 arXiv   pre-print
End-to-end AutoML has attracted intensive interests from both academia and industry, which automatically searches for ML pipelines in a space induced by feature engineering, algorithm/model selection,  ...  We present VolcanoML, a scalable and extensible framework that facilitates systematic exploration of large AutoML search spaces.  ...  In addition, we also compare VolcanoML with four commercial AutoML platforms from Google, Amazon AWS, Microsoft Azure, and Oracle.  ... 
arXiv:2107.08861v1 fatcat:twfsk2u54rhejpc65fs6j35pwi

Preprocessor Selection for Machine Learning Pipelines [article]

Brandon Schoenfeld, Christophe Giraud-Carrier, Mason Poggemann, Jarom Christensen, Kevin Seppi
2018 arXiv   pre-print
Here, we conduct an extensive empirical study over a wide range of learning algorithms and preprocessors, and use metalearning to determine when one should make use of preprocessors in ML pipeline design  ...  In other words, practical solutions consist of pipelines of machine learning operators rather than single algorithms.  ...  The two metalearning agents, Mode and Oracle, perform best.  ... 
arXiv:1810.09942v1 fatcat:cxwm76e57bdh7pzx6mmmyyuvwq

AutonoML: Towards an Integrated Framework for Autonomous Machine Learning [article]

David Jacob Kedziora and Katarzyna Musial and Bogdan Gabrys
2022 arXiv   pre-print
Central to this drive is the appeal of engineering a computational system that both discovers and deploys high-performance solutions to arbitrary ML problems with minimal human interaction.  ...  However, these ambitions are unlikely to be achieved in a robust manner without the broader synthesis of various mechanisms and theoretical frameworks, which, at the present time, remain scattered across  ...  However, SMBOs remain employed as an alternative, with Fast LineAr SearcH (FLASH) as a recent attempt to upgrade Bayesian optimisation for ML pipelines, separating CASH from pipeline search and applying  ... 
arXiv:2012.12600v2 fatcat:6rj4ubhcjncvddztjs7tql3itq

A classification and review of tools for developing and interacting with machine learning systems

Eduardo Mosqueira-Rey, Elena Hernández Pereira, David Alonso-Ríos, José Bobes-Bascarán
2022 Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing  
For this purpose, we suggest a classification of the tools in which the categories are organized following the development lifecycle of an ML system and we make a review of the existing tools within each  ...  approaches to a more engineering perspective and the ability to make it easier to develop intelligent systems for people without an educational background in the area, in order to move the focus from  ...  ACKNOWLEDGMENTS This work has been supported by the State Research Agency of the Spanish Government, grant (PID2019-107194GB-I00 / AEI / 10.13039/501100011033) and by the Xunta de Galicia, grant (ED431C  ... 
doi:10.1145/3477314.3507310 fatcat:htgi7cqacfhsrckw4wcwz6tgsi

Declarative Machine Learning Systems [article]

Piero Molino, Christopher Ré
2021 arXiv   pre-print
These new ML systems will not require users to fully understand all the details of how models are trained and utilized for obtaining predictions.  ...  In the last years machine learning (ML) has moved from a academic endeavor to a pervasive technology adopted in almost every aspect of computing.  ...  Acknowledgements The authors want to thank Antonio Vergari, Karan Goel, Sahaana Suri, Chip Huyen, Dan Fu, Arun Kumar and Michael Cafarella for insightful comments and suggestions.  ... 
arXiv:2107.08148v1 fatcat:fxxfr3romne5vbrdmziwwlzbfy

Automated Deep Learning: Neural Architecture Search Is Not the End [article]

Xuanyi Dong, David Jacob Kedziora, Katarzyna Musial, Bogdan Gabrys
2022 arXiv   pre-print
Deep learning (DL) has proven to be a highly effective approach for developing models in diverse contexts, including visual perception, speech recognition, and machine translation.  ...  In so doing, this work also proposes a comprehensive set of ten criteria by which to assess existing work in both individual publications and broader research areas.  ...  Acknowledgments: XD and DJK acknowledge financial support secured by BG, which funded their participation in this study and the ongoing "Automated and Autonomous Machine Learning" project as part of the  ... 
arXiv:2112.09245v3 fatcat:dujfh7pzmzbrtdyoshkl4kpbsm

A Generic Deep Learning Based Cough Analysis System from Clinically Validated Samples for Point-of-Need Covid-19 Test and Severity Levels

Javier Andreu-Perez, Humberto Perez-Espinosa, Eva Timonet, Mehrin Kiani, Manuel Ivan Giron-Perez, Alma B. Benitez-Trinidad, Delaram Jarchi, Alejandro Rosales, Nick Gkatzoulis, Orion F. Reyes-Galaviz, Alejandro Torres, Carlos Alberto Reyes-Garcia (+2 others)
2021 IEEE Transactions on Services Computing  
Our proposed web tool and underpinning algorithm for the robust, fast, point-of-need identification of Covid-19 facilitates the rapid detection of the infection.  ...  Our proposed generic method is an algorithm based on Empirical Mode Decomposition (EMD) with subsequent classification based on a tensor of audio features and a deep artificial neural network classifier  ...  We are also grateful to the ESRC Impact Acceleration Account (ES/T501815/1) at Essex University for supporting this research, and Oracle Corporation for providing extra computational resources for this  ... 
doi:10.1109/tsc.2021.3061402 fatcat:uwvple3b55f6rkx7h5g7mye2o4

A survey on data‐efficient algorithms in big data era

Amina Adadi
2021 Journal of Big Data  
Then, it presents a comprehensive review of existing data-efficient methods and systematizes them into four categories.  ...  This has triggered a serious debate in both the industrial and academic communities calling for more data-efficient models that harness the power of artificial learners while achieving good results with  ...  Hence, one of the missions of autoML is to automatically manage data quality and quantity in the first step of the pipeline.  ... 
doi:10.1186/s40537-021-00419-9 fatcat:v4uahsvhlzdldlxqf24bshmja4

Metamorphic Testing for Object Detection Systems [article]

Shuai Wang, Zhendong Su
2019 arXiv   pre-print
between the original and synthetic images after excluding the prediction results on the inserted objects.  ...  MetaOD is designed as a streamlined workflow that performs object extraction, selection, and insertion.  ...  Fast-RCNN [27] introduced a modern end-to-end prediction pipeline.  ... 
arXiv:1912.12162v1 fatcat:hbgvau3jfbelhpxjyotag6epze

Predictive Modeling for Optimization of Field Operations in Bike-Sharing Systems

Simon Ruffieux, Elena Mugellini, Omar Abou Khaled
2019 2019 6th Swiss Conference on Data Science (SDS)  
This talk gives an overview of the state of the art in AutoML with a focus on practical applicability in a business context, and provides recent benchmark results on the most important AutoML algorithms  ...  By setting the price dynamically for each transaction, a provider can react fast and directly to a changing environment, hence steer availability and revenue.  ... 
doi:10.1109/sds.2019.00011 dblp:conf/sds2/RuffieuxMK19 fatcat:z5bcj6f2pjczvbwp4zooavn6ve

Extreme Algorithm Selection With Dyadic Feature Representation [article]

Alexander Tornede, Marcel Wever, Eyke Hüllermeier
2020 arXiv   pre-print
We assess the applicability of state-of-the-art AS techniques to the XAS setting and propose approaches leveraging a dyadic feature representation in which both problem instances and algorithms are described  ...  Algorithm selection (AS) deals with selecting an algorithm from a fixed set of candidate algorithms most suitable for a specific instance of an algorithmic problem, e.g., choosing solvers for SAT problems  ...  The optimal selector is called oracle and is defined as s * (i) . .= arg max aA E m(i, a) (1) for all i ∈ I.  ... 
arXiv:2001.10741v1 fatcat:ica4qzrqk5dljodeyic6zwrly4
« Previous Showing results 1 — 15 out of 57 results