{"id":3397,"date":"2023-01-28T01:15:04","date_gmt":"2023-01-28T01:15:04","guid":{"rendered":"https:\/\/www.etran.rs\/2023\/?p=3397"},"modified":"2023-03-05T22:09:58","modified_gmt":"2023-03-05T22:09:58","slug":"gorana-gojic-the-institute-for-artificial-intelligence-research-and-development-of-serbia-novi-sad-serbia-non-adversarial-robustness-of-deep-learning-methods","status":"publish","type":"post","link":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397","title":{"rendered":"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision"},"content":{"rendered":"

Section AI (Artificial Intelligence)<\/b><\/span><\/p>\n

INVITED PAPER<\/b><\/u><\/span>:<\/b><\/span><\/p>\n

Non-adversarial Robustness of Deep Learning Methods for <\/span>Computer Vision<\/span><\/span><\/span><\/p>\n

Gorana Goji\u0107,<\/b><\/span><\/p>\n

The Institute for Artificial Intelligence Research and Development of Serbia,<\/span><\/p>\n

Novi Sad, Serbia<\/span><\/p>\n

\u00a0<\/p>\n

Abstract:<\/b><\/span><\/p>\n

Non-adversarial robustness, also known as natural robustness, is a property of deep learning models that enables them to maintain performance even when faced with distribution shifts caused by natural variations in data. However, achieving this property is challenging because it is difficult to predict in advance the types of distribution shifts that may occur. To address this challenge, researchers have proposed various approaches, some of which anticipate potential distribution shifts, while others utilize knowledge about the shifts that have already occurred to enhance model generalisability. In this paper, we present a brief overview of the most recent techniques for improving the robustness of computer vision methods, as well as a summary of commonly used robustness benchmark datasets for evaluating the model\u2019s performance under data distribution shifts. Finally, we examine the strengths and limitations of the approaches reviewed and identify general trends in deep learning robustness improvement for computer vision.<\/span><\/p>\n

\u00a0<\/p>\n

\u00a0<\/p>\n

Short Bio:<\/b><\/span><\/p>\n\n\n\n
\n

\"\"<\/p>\n<\/td>\n

\n

Gorana Goji\u0107<\/b> received her B.Sc. with honours and M.Sc. degrees in electrical engineering and computing from the University of Novi Sad, Faculty of Technical Sciences, Novi Sad, Serbia, in 2015 and 2016, respectively. She is currently working as a researcher at The Institute for Artificial Intelligence Research and Development of Serbia as a researcher, while pursuing a Ph.D. degree in the field of machine learning at the Faculty of Technical Sciences, University of Novi Sad. <\/span><\/p>\n

Her professional interests include deep learning and computer vision applied in medical image processing.<\/span><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\u00a0<\/p>\n","protected":false},"excerpt":{"rendered":"

Section AI (Artificial Intelligence) INVITED PAPER: Non-adversarial Robustness of Deep Learning Methods for Computer Vision Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia \u00a0 Abstract: Non-adversarial robustness, also known as natural robustness, is a property of deep learning models that enables them to maintain performance even when faced \u2026 <\/p>\n","protected":false},"author":1,"featured_media":3398,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_themeisle_gutenberg_block_has_review":false,"footnotes":""},"categories":[117],"tags":[],"class_list":["post-3397","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-invited-papers"],"yoast_head":"\nGorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision -<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision -\" \/>\n<meta property=\"og:description\" content=\"Section AI (Artificial Intelligence) INVITED PAPER: Non-adversarial Robustness of Deep Learning Methods for Computer Vision Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia \u00a0 Abstract: Non-adversarial robustness, also known as natural robustness, is a property of deep learning models that enables them to maintain performance even when faced \u2026\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397\" \/>\n<meta property=\"article:published_time\" content=\"2023-01-28T01:15:04+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-03-05T22:09:58+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"212\" \/>\n\t<meta property=\"og:image:height\" content=\"256\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Uzi Ramadani\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Uzi Ramadani\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397\"},\"author\":{\"name\":\"Uzi Ramadani\",\"@id\":\"https:\/\/www.etran.rs\/2023\/#\/schema\/person\/622a9955a8e46af42b7de704d8d7f802\"},\"headline\":\"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision\",\"datePublished\":\"2023-01-28T01:15:04+00:00\",\"dateModified\":\"2023-03-05T22:09:58+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397\"},\"wordCount\":300,\"publisher\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/#organization\"},\"image\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg\",\"articleSection\":[\"Invited Papers\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397\",\"url\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397\",\"name\":\"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision -\",\"isPartOf\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg\",\"datePublished\":\"2023-01-28T01:15:04+00:00\",\"dateModified\":\"2023-03-05T22:09:58+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage\",\"url\":\"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg\",\"contentUrl\":\"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg\",\"width\":212,\"height\":256},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.etran.rs\/2023\/en\/home-english\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.etran.rs\/2023\/#website\",\"url\":\"https:\/\/www.etran.rs\/2023\/\",\"name\":\"\",\"description\":\"X IcETRAN Conference 2022 - East Sarajevo,Bosnia and Herzegovina, June 5-8, 2023.\",\"publisher\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.etran.rs\/2023\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.etran.rs\/2023\/#organization\",\"name\":\"ETRAN Society\",\"url\":\"https:\/\/www.etran.rs\/2023\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.etran.rs\/2023\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.etran.rs\/2021\/wp-content\/uploads\/2021\/06\/ETRAN-logo.png\",\"contentUrl\":\"https:\/\/www.etran.rs\/2021\/wp-content\/uploads\/2021\/06\/ETRAN-logo.png\",\"width\":338,\"height\":89,\"caption\":\"ETRAN Society\"},\"image\":{\"@id\":\"https:\/\/www.etran.rs\/2023\/#\/schema\/logo\/image\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.etran.rs\/2023\/#\/schema\/person\/622a9955a8e46af42b7de704d8d7f802\",\"name\":\"Uzi Ramadani\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.etran.rs\/2023\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/8a8990f6e1e9e4daee70e1f7e6e99023e215901a00579a7ba2ecc33b5aba6655?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/8a8990f6e1e9e4daee70e1f7e6e99023e215901a00579a7ba2ecc33b5aba6655?s=96&d=mm&r=g\",\"caption\":\"Uzi Ramadani\"},\"sameAs\":[\"https:\/\/www.etran.rs\/2021\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision -","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:\/\/www.etran.rs\/2023\/en\/archives\/3397","og_locale":"en_US","og_type":"article","og_title":"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision -","og_description":"Section AI (Artificial Intelligence) INVITED PAPER: Non-adversarial Robustness of Deep Learning Methods for Computer Vision Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia \u00a0 Abstract: Non-adversarial robustness, also known as natural robustness, is a property of deep learning models that enables them to maintain performance even when faced \u2026","og_url":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397","article_published_time":"2023-01-28T01:15:04+00:00","article_modified_time":"2023-03-05T22:09:58+00:00","og_image":[{"width":212,"height":256,"url":"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg","type":"image\/jpeg"}],"author":"Uzi Ramadani","twitter_card":"summary_large_image","twitter_misc":{"Written by":"Uzi Ramadani","Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#article","isPartOf":{"@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397"},"author":{"name":"Uzi Ramadani","@id":"https:\/\/www.etran.rs\/2023\/#\/schema\/person\/622a9955a8e46af42b7de704d8d7f802"},"headline":"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision","datePublished":"2023-01-28T01:15:04+00:00","dateModified":"2023-03-05T22:09:58+00:00","mainEntityOfPage":{"@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397"},"wordCount":300,"publisher":{"@id":"https:\/\/www.etran.rs\/2023\/#organization"},"image":{"@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage"},"thumbnailUrl":"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg","articleSection":["Invited Papers"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397","url":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397","name":"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision -","isPartOf":{"@id":"https:\/\/www.etran.rs\/2023\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage"},"image":{"@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage"},"thumbnailUrl":"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg","datePublished":"2023-01-28T01:15:04+00:00","dateModified":"2023-03-05T22:09:58+00:00","breadcrumb":{"@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.etran.rs\/2023\/en\/archives\/3397"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#primaryimage","url":"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg","contentUrl":"https:\/\/www.etran.rs\/2023\/wp-content\/uploads\/2023\/01\/goranagojic.jpg","width":212,"height":256},{"@type":"BreadcrumbList","@id":"https:\/\/www.etran.rs\/2023\/en\/archives\/3397#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.etran.rs\/2023\/en\/home-english"},{"@type":"ListItem","position":2,"name":"Gorana Goji\u0107, The Institute for Artificial Intelligence Research and Development of Serbia, Novi Sad, Serbia : Non-adversarial Robustness of Deep Learning Methods for Computer Vision"}]},{"@type":"WebSite","@id":"https:\/\/www.etran.rs\/2023\/#website","url":"https:\/\/www.etran.rs\/2023\/","name":"","description":"X IcETRAN Conference 2022 - East Sarajevo,Bosnia and Herzegovina, June 5-8, 2023.","publisher":{"@id":"https:\/\/www.etran.rs\/2023\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.etran.rs\/2023\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.etran.rs\/2023\/#organization","name":"ETRAN Society","url":"https:\/\/www.etran.rs\/2023\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.etran.rs\/2023\/#\/schema\/logo\/image\/","url":"https:\/\/www.etran.rs\/2021\/wp-content\/uploads\/2021\/06\/ETRAN-logo.png","contentUrl":"https:\/\/www.etran.rs\/2021\/wp-content\/uploads\/2021\/06\/ETRAN-logo.png","width":338,"height":89,"caption":"ETRAN Society"},"image":{"@id":"https:\/\/www.etran.rs\/2023\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/www.etran.rs\/2023\/#\/schema\/person\/622a9955a8e46af42b7de704d8d7f802","name":"Uzi Ramadani","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.etran.rs\/2023\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/8a8990f6e1e9e4daee70e1f7e6e99023e215901a00579a7ba2ecc33b5aba6655?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/8a8990f6e1e9e4daee70e1f7e6e99023e215901a00579a7ba2ecc33b5aba6655?s=96&d=mm&r=g","caption":"Uzi Ramadani"},"sameAs":["https:\/\/www.etran.rs\/2021"]}]}},"_links":{"self":[{"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/posts\/3397","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/comments?post=3397"}],"version-history":[{"count":3,"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/posts\/3397\/revisions"}],"predecessor-version":[{"id":3463,"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/posts\/3397\/revisions\/3463"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/media\/3398"}],"wp:attachment":[{"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/media?parent=3397"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/categories?post=3397"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.etran.rs\/2023\/wp-json\/wp\/v2\/tags?post=3397"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}