End-to-End Machine Learning for Experimental Physics: Using Simulated Data to Train a Neural Network for Object Detection in Video Microscopy

We demonstrate a method for training a convolutional neural network with simulated images for usage on real-world experimental data. Modern machine learning methods require large, robust training data sets to generate accurate predictions. Generating these large training sets requires a significant...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Minor, Eric N, Howard, Stian D, Green, Adam A. S, Park, Cheol S, Clark, Noel A
Format:	Artikel
Sprache:	eng
Schlagworte:	Physics - Disordered Systems and Neural Networks Physics - Soft Condensed Matter
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

container_end_page
container_issue
container_start_page
container_title
container_volume
creator	Minor, Eric N Howard, Stian D Green, Adam A. S Park, Cheol S Clark, Noel A
description	We demonstrate a method for training a convolutional neural network with simulated images for usage on real-world experimental data. Modern machine learning methods require large, robust training data sets to generate accurate predictions. Generating these large training sets requires a significant up-front time investment that is often impractical for small-scale applications. Here we demonstrate a `full-stack' computational solution, where the training data set is generated on-the-fly using a noise injection process to produce simulated data characteristic of the experimental system.
doi_str_mv	10.48550/arxiv.1908.05271
format	Article
fullrecord	<record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1908_05271</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1908_05271</sourcerecordid><originalsourceid>FETCH-LOGICAL-a671-194b595237c842028928e4b1b88728c27cc606c3f3aed6f1c795029a704eab143</originalsourceid><addsrcrecordid>eNotkEtOwzAURTNhgAoLYMTbQILtfOwwQ235SGmLRGFavTgv1NDaleNCswj2TFoYncG9utI9UXTFWZKpPGc36A_mK-ElUwnLheTn0c_UNnFw8QCYoV4bS1ARemvsO7TOw_SwI2-2ZANu4Hndd0Z3t_DaHfMXs91vMFADEwwIwcHSo7GAMKe9H_pzCt_Of56GFvUH6QATCgOMszAU30xDDmZGe9dpt-svorMWNx1d_nMULe-ny_FjXC0ensZ3VYyF5DEvszovc5FKrTLBhCqFoqzmtVJSKC2k1gUrdNqmSE3Rci3LnIkSJcsIa56lo-j6b_bkY7Ub_qHvV0cvq5OX9Bdxql4d</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>End-to-End Machine Learning for Experimental Physics: Using Simulated Data to Train a Neural Network for Object Detection in Video Microscopy</title><source>arXiv.org</source><creator>Minor, Eric N ; Howard, Stian D ; Green, Adam A. S ; Park, Cheol S ; Clark, Noel A</creator><creatorcontrib>Minor, Eric N ; Howard, Stian D ; Green, Adam A. S ; Park, Cheol S ; Clark, Noel A</creatorcontrib><description>We demonstrate a method for training a convolutional neural network with simulated images for usage on real-world experimental data. Modern machine learning methods require large, robust training data sets to generate accurate predictions. Generating these large training sets requires a significant up-front time investment that is often impractical for small-scale applications. Here we demonstrate a `full-stack' computational solution, where the training data set is generated on-the-fly using a noise injection process to produce simulated data characteristic of the experimental system.</description><identifier>DOI: 10.48550/arxiv.1908.05271</identifier><language>eng</language><subject>Physics - Disordered Systems and Neural Networks ; Physics - Soft Condensed Matter</subject><creationdate>2019-08</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1908.05271$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1908.05271$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Minor, Eric N</creatorcontrib><creatorcontrib>Howard, Stian D</creatorcontrib><creatorcontrib>Green, Adam A. S</creatorcontrib><creatorcontrib>Park, Cheol S</creatorcontrib><creatorcontrib>Clark, Noel A</creatorcontrib><title>End-to-End Machine Learning for Experimental Physics: Using Simulated Data to Train a Neural Network for Object Detection in Video Microscopy</title><description>We demonstrate a method for training a convolutional neural network with simulated images for usage on real-world experimental data. Modern machine learning methods require large, robust training data sets to generate accurate predictions. Generating these large training sets requires a significant up-front time investment that is often impractical for small-scale applications. Here we demonstrate a `full-stack' computational solution, where the training data set is generated on-the-fly using a noise injection process to produce simulated data characteristic of the experimental system.</description><subject>Physics - Disordered Systems and Neural Networks</subject><subject>Physics - Soft Condensed Matter</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotkEtOwzAURTNhgAoLYMTbQILtfOwwQ235SGmLRGFavTgv1NDaleNCswj2TFoYncG9utI9UXTFWZKpPGc36A_mK-ElUwnLheTn0c_UNnFw8QCYoV4bS1ARemvsO7TOw_SwI2-2ZANu4Hndd0Z3t_DaHfMXs91vMFADEwwIwcHSo7GAMKe9H_pzCt_Of56GFvUH6QATCgOMszAU30xDDmZGe9dpt-svorMWNx1d_nMULe-ny_FjXC0ensZ3VYyF5DEvszovc5FKrTLBhCqFoqzmtVJSKC2k1gUrdNqmSE3Rci3LnIkSJcsIa56lo-j6b_bkY7Ub_qHvV0cvq5OX9Bdxql4d</recordid><startdate>20190812</startdate><enddate>20190812</enddate><creator>Minor, Eric N</creator><creator>Howard, Stian D</creator><creator>Green, Adam A. S</creator><creator>Park, Cheol S</creator><creator>Clark, Noel A</creator><scope>GOX</scope></search><sort><creationdate>20190812</creationdate><title>End-to-End Machine Learning for Experimental Physics: Using Simulated Data to Train a Neural Network for Object Detection in Video Microscopy</title><author>Minor, Eric N ; Howard, Stian D ; Green, Adam A. S ; Park, Cheol S ; Clark, Noel A</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a671-194b595237c842028928e4b1b88728c27cc606c3f3aed6f1c795029a704eab143</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Physics - Disordered Systems and Neural Networks</topic><topic>Physics - Soft Condensed Matter</topic><toplevel>online_resources</toplevel><creatorcontrib>Minor, Eric N</creatorcontrib><creatorcontrib>Howard, Stian D</creatorcontrib><creatorcontrib>Green, Adam A. S</creatorcontrib><creatorcontrib>Park, Cheol S</creatorcontrib><creatorcontrib>Clark, Noel A</creatorcontrib><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Minor, Eric N</au><au>Howard, Stian D</au><au>Green, Adam A. S</au><au>Park, Cheol S</au><au>Clark, Noel A</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>End-to-End Machine Learning for Experimental Physics: Using Simulated Data to Train a Neural Network for Object Detection in Video Microscopy</atitle><date>2019-08-12</date><risdate>2019</risdate><abstract>We demonstrate a method for training a convolutional neural network with simulated images for usage on real-world experimental data. Modern machine learning methods require large, robust training data sets to generate accurate predictions. Generating these large training sets requires a significant up-front time investment that is often impractical for small-scale applications. Here we demonstrate a `full-stack' computational solution, where the training data set is generated on-the-fly using a noise injection process to produce simulated data characteristic of the experimental system.</abstract><doi>10.48550/arxiv.1908.05271</doi><oa>free_for_read</oa></addata></record>
fulltext	fulltext_linktorsrc
identifier	DOI: 10.48550/arxiv.1908.05271
ispartof
issn
language	eng
recordid	cdi_arxiv_primary_1908_05271
source	arXiv.org
subjects	Physics - Disordered Systems and Neural Networks Physics - Soft Condensed Matter
title	End-to-End Machine Learning for Experimental Physics: Using Simulated Data to Train a Neural Network for Object Detection in Video Microscopy
url	https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-22T07%3A05%3A37IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=End-to-End%20Machine%20Learning%20for%20Experimental%20Physics:%20Using%20Simulated%20Data%20to%20Train%20a%20Neural%20Network%20for%20Object%20Detection%20in%20Video%20Microscopy&rft.au=Minor,%20Eric%20N&rft.date=2019-08-12&rft_id=info:doi/10.48550/arxiv.1908.05271&rft_dat=%3Carxiv_GOX%3E1908_05271%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true