{"id":2813,"date":"2022-11-16T06:48:14","date_gmt":"2022-11-16T05:48:14","guid":{"rendered":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/?p=2813"},"modified":"2022-11-21T12:58:12","modified_gmt":"2022-11-21T11:58:12","slug":"podcast-few-data-learning","status":"publish","type":"post","link":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/","title":{"rendered":"Podcast: Few Data Learning"},"content":{"rendered":"

Anwendung von Machine Learning Verfahren trotz geringer Datenbasis<\/strong><\/p>\n

<\/p>\n

In der hier angek\u00fcndigten Podcastfolge spreche ich mit Dr. Christian Menden \u00fcber die Methoden und Anwendungsm\u00f6glichkeiten seiner Kompetenzs\u00e4ule \u00bbFew Data Learning<\/a>\u00ab.
\nIn dieser Kompetenzs\u00e4ule geht es darum, eine kleine oder qualitativ unzureichende Datenbasis anzureichern, damit trotzdem Machine Learning Verfahren eingesetzt werden k\u00f6nnen. \u00bbFew Data Learning\u00ab besch\u00e4ftigt sich also damit, Daten f\u00fcr KI-Verfahren \u00fcberhaupt erst nutzbar zu machen.<\/p>\n

Was ist Few Data Learning?
\n<\/strong>Ein wichtiger Baustein f\u00fcr Machine Learning-Verfahren ist das Vorhandensein einer gro\u00dfen Menge von qualitativ hochwertigen Daten. In der Praxis gibt es allerdings F\u00e4lle, in denen einfach nicht gen\u00fcgend Daten vorhanden sind oder es viele fehlerhafte Daten gibt, wenn beispielsweise in einer Fertigung einzelne Sensoren ausfallen oder Anlagen auf ein neues Produkt umger\u00fcstet werden. In manchen Anwendungsbereichen ist es auch einfach so, dass die Annotation und Strukturierung von Daten sehr aufw\u00e4ndig ist und nicht vollautomatisiert m\u00f6glich ist, wie bei Gewebescans im medizinischen Bereich<\/a>.<\/p>\n

Unterschiedliche Verfahren f\u00fcr unzureichende Datens\u00e4tze
\n<\/strong>In verschiedenen Verfahren k\u00f6nnen Datens\u00e4tze angereichert werden. Fehlen in einem Datensatz einige wenige Daten Punkte, werden L\u00fccken im Imputations-Verfahren gef\u00fcllt.
\nSind die L\u00fccken aber gr\u00f6\u00dfer und substantiell, werden die wenigen vorhandenen Daten genutzt und nach Mustern gesucht, um zus\u00e4tzliche Daten zu generieren. Sind gar keine Daten vorhanden, k\u00f6nnen Simulationstechniken eingesetzt werden, um Datens\u00e4tze zu generieren.
\nDie Einsatzm\u00f6glichkeiten in unterschiedlichsten Bereichen ist riesig: von der Qualit\u00e4tskontrolle von Produkten, Umr\u00fcstung einer bestehenden Fertigung f\u00fcr ein neues Produkt, bis hin zur bildgest\u00fctzten Diagnose von Darmkrebsregionen.<\/p>\n

Welche Methoden in welcher Anwendung eingesetzt werden und was Few Data Learning mit Few Labels Learning<\/a> zu tun hat, verr\u00e4t Ihnen Christian in der neuen Folge!<\/p>\nhttps:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-content\/uploads\/2022\/11\/K9_Few_Data_Learning_ADA.mp3<\/a><\/audio>\n

<\/p>\n

Using machine learning methods despite a small database<\/h3>\n

In this podcast episode I talk with Dr. Christian Menden about the methods and possible applications of his competence pillar Few Data Learning.<\/a><\/strong><\/p>\n

This competence pillar is about enriching a small or qualitatively insufficient data base so that Machine Learning methods can still be used. “Few Data Learning” is therefore a research area that deals with making data usable for AI procedures<\/p>\n

What is Few Data Learning?
\n<\/strong>An important foundation for machine learning methods is the presence of a large amount of high-quality data. In practice, there are cases where there simply isn’t enough data or there is a lot of erroneous data, such as when individual sensors fail in a manufacturing facility or equipment is being converted to create a new product. In some application areas, it is also simply the case that the annotation and structuring of data is very time-consuming and cannot be fully automated, as in the case of tissue scans in the medical field<\/a>.<\/p>\n

Different procedures for insufficient data sets
\n<\/strong>Data sets can be enriched in different procedures:
\n– a few data points are missing in a data set, gaps are filled using the imputation procedure
\n– the gaps are larger and substantial, the few existing data are used and patterns are searched to generate additional data
\n– no data are available at all, simulation techniques can be used to generate data sets<\/p>\n

The potential applications in a wide variety of fields are enormous: from quality control of products, to production of different products in an existing manufacturing facility, to image-based diagnosis of colon cancer regions.<\/p>\n

Which methods are used in which application and what Few Data Learning has to do with Few Labels Learning<\/a>, Christian reveals in the new episode!<\/p>\n

https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-content\/uploads\/2022\/10\/ADA_Lovelace_Center_Few_Data_Learning.mp4<\/a><\/video><\/div>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n","protected":false},"excerpt":{"rendered":"

Anwendung von Machine Learning Verfahren trotz geringer Datenbasis<\/p>\n","protected":false},"author":2,"featured_media":2816,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[54,1,55],"tags":[105,80,67,110,64],"class_list":["post-2813","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-anwendungen","category-forschung","category-menschen","tag-adabloggt","tag-adawillswissen","tag-ai","tag-few-data-learning","tag-ki"],"acf":[],"yoast_head":"\nPodcast: Few Data Learning - adalovelacecenter-blog<\/title>\n<meta name=\"description\" content=\"Podcast mit Dr. Christian Menden zur Kompetenzs\u00e4ule \u00bbFew Data Learning\u00ab: Anwendung von Machine Learning Verfahren trotz geringer Datenbasis\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/\" \/>\n<meta property=\"og:locale\" content=\"de_DE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Anwendung von Machine Learning Verfahren trotz geringer Datenbasis\" \/>\n<meta property=\"og:description\" content=\"In der neuen Podcastfolge der Serie \u00bbAda wills wissen\u00ab erfahrt ihr mehr \u00fcber Few Data Learning, Methoden und spannende Anwendungsgebiete in der Fertigungsindustrie oder dem Medizinbereich.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/\" \/>\n<meta property=\"og:site_name\" content=\"adalovelacecenter-blog\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/de-de.facebook.com\/FraunhoferIIS\" \/>\n<meta property=\"article:published_time\" content=\"2022-11-16T05:48:14+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2022-11-21T11:58:12+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-content\/uploads\/2022\/10\/Few-Data-Learning.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1501\" \/>\n\t<meta property=\"og:image:height\" content=\"501\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Anik\u00f3 Enderlein\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:title\" content=\"Neue Podcastfolge: Few Data Learning\" \/>\n<meta name=\"twitter:description\" content=\"Warum das Vorhandensein von Daten nicht immer ausreichend ist, um KI-Verfahren gewinnbringend einzusetzen und wie sehr kleine oder fehlerhafte Datens\u00e4tze dennoch genutzt werden k\u00f6nnen, erkl\u00e4rt euch Dr. Christian Menden in der neuen \u00bbAda wills wissen\u00ab-Podcastfolge.\" \/>\n<meta name=\"twitter:image\" content=\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-content\/uploads\/2022\/10\/Few-Data-Learning.png\" \/>\n<meta name=\"twitter:creator\" content=\"@FraunhoferIIS\" \/>\n<meta name=\"twitter:site\" content=\"@FraunhoferIIS\" \/>\n<meta name=\"twitter:label1\" content=\"Verfasst von\" \/>\n\t<meta name=\"twitter:data1\" content=\"Anik\u00f3 Enderlein\" \/>\n\t<meta name=\"twitter:label2\" content=\"Gesch\u00e4tzte Lesezeit\" \/>\n\t<meta name=\"twitter:data2\" content=\"4\u00a0Minuten\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/\",\"url\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/\",\"name\":\"Podcast: Few Data Learning - adalovelacecenter-blog\",\"isPartOf\":{\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#website\"},\"datePublished\":\"2022-11-16T05:48:14+00:00\",\"dateModified\":\"2022-11-21T11:58:12+00:00\",\"author\":{\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#\/schema\/person\/db1f6550f5aeffe2718604493fee7639\"},\"description\":\"Podcast mit Dr. Christian Menden zur Kompetenzs\u00e4ule \u00bbFew Data Learning\u00ab: Anwendung von Machine Learning Verfahren trotz geringer Datenbasis\",\"breadcrumb\":{\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/#breadcrumb\"},\"inLanguage\":\"de\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Startseite\",\"item\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Podcast: Few Data Learning\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#website\",\"url\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/\",\"name\":\"adalovelacecenter-blog\",\"description\":\"Kompetenzzentrum f\u00fcr Analytics, Daten und Applikationen\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"de\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#\/schema\/person\/db1f6550f5aeffe2718604493fee7639\",\"name\":\"Anik\u00f3 Enderlein\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/942081197915d7b4a2d8f7bfc7b23d11dd4ecb2dfd82459877ad80d63ab5cc9e?s=96&d=blank&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/942081197915d7b4a2d8f7bfc7b23d11dd4ecb2dfd82459877ad80d63ab5cc9e?s=96&d=blank&r=g\",\"caption\":\"Anik\u00f3 Enderlein\"},\"url\":\"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/author\/aniko_enderlein\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Podcast: Few Data Learning - adalovelacecenter-blog","description":"Podcast mit Dr. Christian Menden zur Kompetenzs\u00e4ule \u00bbFew Data Learning\u00ab: Anwendung von Machine Learning Verfahren trotz geringer Datenbasis","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/","og_locale":"de_DE","og_type":"article","og_title":"Anwendung von Machine Learning Verfahren trotz geringer Datenbasis","og_description":"In der neuen Podcastfolge der Serie \u00bbAda wills wissen\u00ab erfahrt ihr mehr \u00fcber Few Data Learning, Methoden und spannende Anwendungsgebiete in der Fertigungsindustrie oder dem Medizinbereich.","og_url":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/","og_site_name":"adalovelacecenter-blog","article_publisher":"https:\/\/de-de.facebook.com\/FraunhoferIIS","article_published_time":"2022-11-16T05:48:14+00:00","article_modified_time":"2022-11-21T11:58:12+00:00","og_image":[{"width":1501,"height":501,"url":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-content\/uploads\/2022\/10\/Few-Data-Learning.png","type":"image\/png"}],"author":"Anik\u00f3 Enderlein","twitter_card":"summary_large_image","twitter_title":"Neue Podcastfolge: Few Data Learning","twitter_description":"Warum das Vorhandensein von Daten nicht immer ausreichend ist, um KI-Verfahren gewinnbringend einzusetzen und wie sehr kleine oder fehlerhafte Datens\u00e4tze dennoch genutzt werden k\u00f6nnen, erkl\u00e4rt euch Dr. Christian Menden in der neuen \u00bbAda wills wissen\u00ab-Podcastfolge.","twitter_image":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-content\/uploads\/2022\/10\/Few-Data-Learning.png","twitter_creator":"@FraunhoferIIS","twitter_site":"@FraunhoferIIS","twitter_misc":{"Verfasst von":"Anik\u00f3 Enderlein","Gesch\u00e4tzte Lesezeit":"4\u00a0Minuten"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/","url":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/","name":"Podcast: Few Data Learning - adalovelacecenter-blog","isPartOf":{"@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#website"},"datePublished":"2022-11-16T05:48:14+00:00","dateModified":"2022-11-21T11:58:12+00:00","author":{"@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#\/schema\/person\/db1f6550f5aeffe2718604493fee7639"},"description":"Podcast mit Dr. Christian Menden zur Kompetenzs\u00e4ule \u00bbFew Data Learning\u00ab: Anwendung von Machine Learning Verfahren trotz geringer Datenbasis","breadcrumb":{"@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/#breadcrumb"},"inLanguage":"de","potentialAction":[{"@type":"ReadAction","target":["https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/podcast-few-data-learning\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Startseite","item":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/"},{"@type":"ListItem","position":2,"name":"Podcast: Few Data Learning"}]},{"@type":"WebSite","@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#website","url":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/","name":"adalovelacecenter-blog","description":"Kompetenzzentrum f\u00fcr Analytics, Daten und Applikationen","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"de"},{"@type":"Person","@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#\/schema\/person\/db1f6550f5aeffe2718604493fee7639","name":"Anik\u00f3 Enderlein","image":{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/942081197915d7b4a2d8f7bfc7b23d11dd4ecb2dfd82459877ad80d63ab5cc9e?s=96&d=blank&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/942081197915d7b4a2d8f7bfc7b23d11dd4ecb2dfd82459877ad80d63ab5cc9e?s=96&d=blank&r=g","caption":"Anik\u00f3 Enderlein"},"url":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/author\/aniko_enderlein\/"}]}},"_links":{"self":[{"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/posts\/2813","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/comments?post=2813"}],"version-history":[{"count":11,"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/posts\/2813\/revisions"}],"predecessor-version":[{"id":2882,"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/posts\/2813\/revisions\/2882"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/media\/2816"}],"wp:attachment":[{"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/media?parent=2813"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/categories?post=2813"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/websites.fraunhofer.de\/adalovelacecenter-blog\/wp-json\/wp\/v2\/tags?post=2813"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}