Filters








87 Hits in 4.0 sec

Towards Evaluating Exploratory Model Building Process with AutoML Systems [article]

Sungsoo Ray Hong, Sonia Castelo, Vito D'Orazio, Christopher Benthune, Aecio Santos, Scott Langevin, David Jonker, Enrico Bertini, Juliana Freire
2020 arXiv   pre-print
While rigorously evaluating how end-users interact with AutoML is crucial, establishing a robust evaluation methodology for such exploratory systems is challenging.  ...  The use of Automated Machine Learning (AutoML) systems are highly open-ended and exploratory.  ...  an end-user's exploratory ML model building process.  ... 
arXiv:2009.00449v1 fatcat:l2giagcq7zhifbwpbwkkgk3gke

Towards Human Centered AutoML [article]

Florian Pfisterer, Janek Thomas, Bernd Bischl
2019 arXiv   pre-print
AutoML systems are currently rising in popularity, as they can build powerful models without human oversight.  ...  While data scientists are often forced to spend the majority of the time available for a given project on data cleaning and exploratory analysis, the time available to practitioners to build actual models  ...  These toolboxes serve as a first step towards making machine-learning accessible to a wide audience of practitioners and build the foundation of most state-of-the-art AutoML systems.  ... 
arXiv:1911.02391v1 fatcat:pvy5i5xvvzbh3kafobjv3pbxci

Visual Analytics for Automated Model Discovery [article]

Dylan Cashman , Kendall Park , Abigail Mosca , Remco Chang University of Wisconsin - Madison,
2018 arXiv   pre-print
However, while autoML tools give users access to arbitrarily complex models, they typically return those models with little context or explanation.  ...  tuning process.  ...  The autoML system can also respond to inquiries about its internal state. For example, when training a complex model, an autoML systems could respond with its status towards completing the task.  ... 
arXiv:1809.10782v2 fatcat:bmpxyqwoabajdkq6o3snmt3dhq

Fits and Starts: Enterprise Use of AutoML and the Role of Humans in the Loop [article]

Anamaria Crisan, Brittany Fiore-Gartland
2021 arXiv   pre-print
Our investigation also captures how data visualization is used in conjunction with AutoML systems.  ...  In this study, we conduct interviews with 29 individuals from organizations of different sizes to characterize how they currently use, or intend to use, AutoML systems in their data science work.  ...  ACKNOWLEDGMENTS The authors wish to acknowledge and thank to study participants for sharing their insights with us.  ... 
arXiv:2101.04296v1 fatcat:rawglgau7fhgvj2npii7if6qwq

An Automated Machine Learning Framework in Unmanned Aircraft Systems: New Insights into Agricultural Management Practices Recognition Approaches

Kai-Yun Li, Niall G. Burnside, Raul Sampaio de Lima, Miguel Villoslada Peciña, Karli Sepp, Victor Henrique Cabral Pinheiro, Bruno Rucy Carneiro Alves de Lima, Ming-Der Yang, Ants Vain, Kalev Sepp
2021 Remote Sensing  
However, a current knowledge gap lies in the integration of AutoML technology and unmanned aircraft systems (UAS) within image-based data classification tasks.  ...  and hyperparameter optimization of the deployable pipeline model for unraveling substance problems.  ...  In addition, this work was supported by Kuusiku Variety Testing Centre, Estonia, with the tests carried out by order of Agricultural Board.  ... 
doi:10.3390/rs13163190 fatcat:sfuuwbpbtjcjdh54yr3yr6lxmq

AutoML to Date and Beyond: Challenges and Opportunities [article]

Shubhra Kanti Karmaker Santu, Md. Mahadi Hassan, Micah J. Smith, Lei Xu, ChengXiang Zhai, Kalyan Veeramachaneni
2021 arXiv   pre-print
These steps often require a prolonged back-and-forth that makes this process inefficient for domain experts and data scientists alike, and keeps so-called AutoML systems from being truly automatic.  ...  In this review article, we introduce a new classification system for AutoML systems, using a seven-tiered schematic to distinguish these systems based on their level of autonomy.  ...  SMBO first builds a model and uses it to determine candidate configuration of hyperparameters; then, it evaluates the loss and updates the model.  ... 
arXiv:2010.10777v4 fatcat:arixmky6erdvhnmboe2sfgbb7a

Automating Data Science: Prospects and Challenges [article]

Tijl De Bie, Luc De Raedt, José Hernández-Orallo, Holger H. Hoos, Padhraic Smyth, Christopher K. I. Williams
2021 arXiv   pre-print
. * Important parts of data science are already being automated, especially in the modeling stages, where techniques such as automated machine learning (AutoML) are gaining traction. * Other aspects are  ...  Given the complexity of typical data science projects and the associated demand for human expertise, automation has the potential to transform the data science process.  ...  However, 'internal' evaluation, which is usually coupled with model building or carried out immediately after, is done in the lab, trying to maximize some metric on held-out data.  ... 
arXiv:2105.05699v1 fatcat:zmfdufmnzrc47hanwx5t33efwu

HyperTendril: Visual Analytics for User-Driven Hyperparameter Optimization of Deep Neural Networks [article]

Heungseok Park, Yoonsoo Nam, Ji-Hoon Kim, Jaegul Choo
2020 arXiv   pre-print
In response, we propose HyperTendril, a web-based visual analytics system that supports user-driven hyperparameter tuning processes in a model-agnostic environment.  ...  We present the evaluation demonstrating how HyperTendril helps users steer their tuning processes via a longitudinal user study based on the analysis of interaction logs and in-depth interviews while we  ...  The generated model data by AutoML processes and the scalar data of the models from backend are passed to the interface with JSON format.  ... 
arXiv:2009.02078v2 fatcat:nc7qijknpnd3xbakwlhlo3vkfa

A Survey on Proactive Customer Care: Enabling Science and Steps to Realize it [article]

Viswanath Ganapathy, Sauptik Dhar, Olimpiya Saha, Pelin Kurt Garberson, Javad Heydari, Mohak Shah
2021 arXiv   pre-print
research and model deployment.  ...  We highlight how such a step-wise approach can be advantageous for accurate model building and helpful for gaining insights into predictive maintenance of electromechanical appliances.  ...  We also acknowledge the Google Cloud Platform (GCP) team for providing us with the computing resources for the analyzing the Blackblaze dataset.  ... 
arXiv:2110.05015v1 fatcat:wjlrvk4hhbhkzlnkokz4pvwzn4

SIMON: Open-Source Knowledge Discovery Platform

Adriana Tomic, Ivan Tomic, Levi Waldron, Ludwig Geistlinger, Max Kuhn, Rachel L. Spreng, Lindsay C. Dahora, Kelly E. Seaton, Georgia Tomaras, Jennifer Hill, Niharika A. Duggal, Ross D. Pollock (+6 others)
2021 Patterns  
Data analysis and knowledge discovery has become more and more important in biology and medicine with the increasing complexity of biological datasets, but the necessarily sophisticated programming skills  ...  With an easy-to-use graphical user interface, standardized pipelines, and automated approach for machine learning and other statistical analysis methods, SIMON helps to identify optimal algorithms and  ...  an appropriate algorithm, optimization of model hyperparameters, and evaluation of the best-performing models. 61 AutoML improves the efficiency of the ML process, and the resulting models often outperform  ... 
doi:10.1016/j.patter.2020.100178 pmid:33511368 pmcid:PMC7815964 fatcat:2itoskc6e5h6hpm56lsbr5abpy

Curated Database and Preliminary AutoML QSAR Model for 5-HT1A Receptor

Natalia Czub, Adam Pacławski, Jakub Szlęk, Aleksander Mendyk
2021 Pharmaceutics  
Furthermore, the predictive model was developed using automated machine learning (AutoML) methods.  ...  Usage of this model in screening processes can significantly improve the process of discovery of new drugs in the field of mental diseases and anticancer therapy.  ...  Serotonin interaction with various receptors leads to modulation of numerous physiological and pathological processes, in both the peripheral and the central nervous system (CNS).  ... 
doi:10.3390/pharmaceutics13101711 pmid:34684004 fatcat:luxz3bkoujggto3dt5qzdb3q3a

Advancing Artificial Intelligence and Machine Learning in the U.S. Government Through Improved Public Competitions [article]

Ezekiel J. Maier
2021 arXiv   pre-print
easier participation, empower deployment of validated models, and incentivize engagement from broad sections of the population.  ...  To address these barriers, enabling the use of public competitions for accelerating AI and ML practice, the U.S. government must leverage methods that protect sensitive data while enabling modelling, enable  ...  Food and Drug Administration (FDA), through its Technology Modernization Action Plan (TMAP) and Data Modernization Action Plan (DMAP), is upgrading FDA's technical infrastructure; building processes for  ... 
arXiv:2112.01275v1 fatcat:p5qimno2bjcznflmq7l7t4heqa

Enabling collaborative data science development with the Ballet framework [article]

Micah J. Smith, Jürgen Cito, Kelvin Lu, Kalyan Veeramachaneni
2021 arXiv   pre-print
While the open-source software development model has led to successful large-scale collaborations in building software systems, data science projects are frequently developed by individuals or small teams  ...  We leverage Ballet to conduct a case study analysis of an income prediction problem with 27 collaborators, and discuss implications for future designers of collaborative projects.  ...  INTRODUCTION The open-source software development model has led to successful, large-scale collaborations in building software libraries, software systems, chess engines, scientific analyses, and more  ... 
arXiv:2012.07816v4 fatcat:dkyxx6wyrbg3vfbff5xc6ipoge

A Continuous Cuffless Blood Pressure Estimation Using Tree-Based Pipeline Optimization Tool

Suliman Mohamed Fati, Amgad Muneer, Nur Arifin Akbar, Shakirah Mohd Taib
2021 Symmetry  
Finally, the proposed approach was evaluated and validated using the mean absolute error (MAE).  ...  As a pre-processing stage, notch filter, band-pass filter, and zero phase filtering were applied by TPOT to eliminate any potential noise inherent in the signal.  ...  Therefore, this paper aims to build an automated machine learning model using TPOT to estimate the blood pressure from the photoplethysmography signals.  ... 
doi:10.3390/sym13040686 fatcat:xseyj6xc4rclfehs4pswq26ose

SystemDS: A Declarative Machine Learning System for the End-to-End Data Science Lifecycle [article]

Matthias Boehm, Iulian Antonov, Sebastian Baunsgaard, Mark Dokter, Robert Ginthoer, Kevin Innerebner, Florijan Klezin, Stefanie Lindstaedt, Arnab Phani, Benjamin Rath, Berthold Reinwald, Shafaq Siddiqi, Sebastian Benjamin Wrede
2020 arXiv   pre-print
However, the data science process is exploratory, and deals with underspecified objectives and a wide variety of heterogeneous data sources.  ...  These systems focus primarily on efficient model training and scoring.  ...  tuning, model debugging, rules and AutoML with domain-specific extensions (e.g., constraints and simulation models).  ... 
arXiv:1909.02976v2 fatcat:hdd36ca7jze7figqzaybfgdqra
« Previous Showing results 1 — 15 out of 87 results