[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117463-en":3,"doc-seo-117463-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},117463,1374391975076,"Riley","https://ap-avatar.wpscdn.com/avatar/14000253ca4ec9f6853?x-image-process=image/resize,m_fixed,w_180,h_180&k=1783305029341752051",8,"Research & Report","Image Watermarking for Machine Learning Datasets","Machine learning success depends on the availability of sufficient training data, yet dataset collection is often slow, costly, and sometimes of limited quality. This creates incentives for illegal use and resale, making data ownership protection increasingly important. The study applies well-known image watermarking methods to machine-learning classification datasets, using an SVD-based scheme. Experiments evaluate robustness, demonstrating good imperceptibility and resilience to update, zero-out, and insertion attacks, while deletion attacks are not successfully addressed.","Image Watermarking for Machine Learning Datasets  \nPalle Maesen  \nCyber Security Group, Delft University of Technology The Netherlands  \n[P.Maesen@student.tudelft.nl](P.Maesen@student.tudelft.nl)  \nDevriş İşler  \nIMDEA Networks Institute & UC3M Spain [devris.isler@imdea.org](devris.isler@imdea.org)  \nNikolaos Laoutaris IMDEA Networks Institute  \nSpain [nikolaos.laoutaris@imdea.org](nikolaos.laoutaris@imdea.org)  \nZekeriya Erkin  \nCyber Security Group, Delft University of Technology The Netherlands  \n[Z.Erkin@tudelft.nl](Z.Erkin@tudelft.nl)  \nABSTRACT  \nMachine learning has received increasing attention for the last decade due to its significant success in classification problems in almost every application domain. For its success, the amount of available data for training plays a crucial role in the creation of a machine-learning model. However, the data-gathering process for machine learning algorithms is a tedious and time-consuming task. In many cases, the developers rely on publicly available datasets, which are not always of high quality. Recently, we are witnessing a data market paradigm where valuable datasets are sold. Thus, once the dataset is created or bought, protecting the dataset against illegal use or (re)sale and establishing intellectual property rights is necessary. In this paper, we investigate the question of deploying wellstudied image watermarking techniques to be applied to classification algorithm datasets, without degrading the quality of the dataset. We investigate whether Singular Value Decomposition (SVD)-based techniques from image watermarking could be deployed on machine learning datasets or not. To this end, we chose the watermarking technique described in [8] and applied it to a machine-learning dataset. We provide experimental results on the robustness of the scheme. Our results show that the watermark embedding scheme provides decent imperceptibility and robustness against update, zeroout, and insertion attacks but, it is not successful against deletion attacks. We believe our work can inspire researchers who might want to consider applying well-studied image watermarking techniques to machine learning datasets.  \nCCS CONCEPTS  \n• Social and professional topics → Computing / technology policy;  \n• Intellectual property → Digital rights management .  \nKEYWORDS  \nWatermarking, machine learning, data ownership, data economy  \nACM Reference Format:  \nPalle Maesen, Devriş İşler, Nikolaos Laoutaris, and Zekeriya Erkin. 2023. Image Watermarking for Machine Learning Datasets. In SecondACM Data  \nThis work is licensed under a Creative Commons Attribution International 4 .0 License.  \nDEC’23, June 18, 2023, Seattle, WA, USA © 2023 Copyright held by the owner/author(s) .  \nACM ISBN 979-8-4007-0846-6/23/06 .  \n[https://doi.org/10.1145/3600046.3600048](https://doi.org/10.1145/3600046.3600048)  \nEconomy Workshop (DEC ’23), June 18, 2023, Seattle, WA, USA. ACM, New York, NY, USA, 7 pages. [https://doi.org/10.1145/3600046.3600048](https://doi.org/10.1145/3600046.3600048)  \n1 INTRODUCTION  \nThe amount of data needed to create state-of-the-art machine learning (ML) models is continuously increasing as the complexity and volumes of these models grow [12] . However, public datasets available for training are limited and sometimes not of high quality. Furthermore, application-specific datasets (e.g., health data) are very valuable. Imagine a company that invested an incredible amount of time and resources into the data-collection/generation. Later, the company sells/shares the dataset [4] in order to have a financial gain (or for any other non-economic incentives such as a public dataset) .1 Therefore, it is essential for companies to protect their ownership of their datasets from infringement, piracy, and/or another entity re-selling their ML datasets without their authorization.  \nWatermarking is a well-known technique for protecting ownership rights in media such as image, video, and audio, database, and machi","cbCaifDzzbiWPaMi","https://ap.wps.com/l/cbCaifDzzbiWPaMi","pdf",885604,1,7,"English","en",105,"# Abstract\n# Introduction\n## Dataset ownership and data economy\n## Watermarking fundamentals and extraction\n## Applying image watermarking to ML datasets\n# Approach","[{\"question\":\"Why is protecting machine-learning datasets important?\",\"answer\":\"Dataset values increase as organizations invest in collection and generation. Once datasets are created or purchased, unauthorized use or resale can occur without proper intellectual property protection.\"},{\"question\":\"What watermarking technique is investigated for ML datasets?\",\"answer\":\"The paper investigates using an SVD-based image watermarking technique and applies it to a machine-learning dataset to enable proof of ownership.\"},{\"question\":\"How robust is the watermarking scheme against attacks?\",\"answer\":\"Experimental results show decent imperceptibility and robustness against update, zero-out, and insertion attacks, but the scheme is not successful against deletion attacks.\"}]","Image Watermarking for Machine Learning Datasets | PDF",1785675975,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"image-watermarking-for-machine-learning-datasets","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/image-watermarking-for-machine-learning-datasets/117463/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is protecting machine-learning datasets important?","Question",{"text":75,"@type":76},"Dataset values increase as organizations invest in collection and generation. Once datasets are created or purchased, unauthorized use or resale can occur without proper intellectual property protection.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What watermarking technique is investigated for ML datasets?",{"text":80,"@type":76},"The paper investigates using an SVD-based image watermarking technique and applies it to a machine-learning dataset to enable proof of ownership.",{"name":82,"@type":73,"acceptedAnswer":83},"How robust is the watermarking scheme against attacks?",{"text":84,"@type":76},"Experimental results show decent imperceptibility and robustness against update, zero-out, and insertion attacks, but the scheme is not successful against deletion attacks.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]