[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119018-en":3,"doc-seo-119018-105":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},119018,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Designing for Deployable, Secure, and Generic Machine Learning Systems - Dissertation Abstract","Machine learning systems enable many image-centric applications but still face persistent design complexity and security vulnerabilities in deep neural networks, while deployment of deep neural network models remains difficult. The dissertation proposes a multimedia prototyping framework for visual analytics that improves reusability of video analysis software with minimal performance overhead. It also introduces robust image-processing methods to resist adversarial perturbations and a compression approach to support deployment constraints. First, the vText framework treats video as text for flexible analysis and linking codecs with vision algorithms, achieving comparable runtime.","Portland State University  \nPDXScholar  \n\n| Dissertations and Theses | Dissertations and Theses |\n| --- | --- |\n| 4-25-2024\u003Cbr>Designing for Deployable, Secure, and Generic Machine Learning Systems\u003Cbr>Li-Yun Wang\u003Cbr>Portland State University\u003Cbr>Follow this and additional works at: [https://pdxscholar.library.pdx.edu/open_access_etds](https://pdxscholar.library.pdx.edu/open_access_etds)\u003Cbr> Part of the Computer Sciences Commons\u003Cbr>Let us know how access to this document benefits you. |  |\n\nRecommended Citation  \nWang, Li-Yun, \"Designing for Deployable, Secure, and Generic Machine Learning Systems\" (2024) . Dissertations and Theses. Paper 6616.  \n[https://doi.org/10.15760/etd.3748](https://doi.org/10.15760/etd.3748)  \nThis Dissertation is brought to you for free and open access. It has been accepted for inclusion in Dissertationsand Theses by an authorized administrator of PDXScholar. Please contact us if we can make this document more accessible: [pdxscholar@pdx.edu](pdxscholar@pdx.edu).  \nDesigning for Deployable, Secure, and Generic Machine Learning Systems  \nby  \nLi-Yun Wang  \nA dissertation submitted in partial fulfillment of the requirements for the degree of  \nDoctor of Philosophy  \nin  \nComputer Science  \nDissertation Committee:  \nWu-chi Feng, Chair  \nBanafsheh Rekabdar  \nAtul Ingle  \nFeng Liu  \nPortland State University  \n2024  \n© 2024 Li-Yun Wang  \ni  \nAbstract  \nMachine learning systems have catalyzed numerous image-centric applications owing to the significant achievements of machine learning algorithms and models. While these systems have showcased the efficacy of machine learning models, certain challenges persist, such as machine learning system design and security vulnerabilities inherent in deep neural networks. Moreover, the deployment of deep neural network models remains a significant hurdle. This dissertation introduces a multimedia prototyping framework tailored for visual analytical applications, improving the reusability of video analysis software tools with minimal performance overhead. Furthermore, we present novel image-processing techniques designed to bolster the robustness of deep neural networks and propose an innovative compression technique to address deployment challenges.  \nFirst, we propose a new software prototyping framework called Video as Text (vText) that analyzes and manipulates the video data as trivial as we handle text data in most Unix and Linux systems to tackle the reusability issue in the existing video analysis tools. The vText paradigm seeks to mimic such programs. We demonstrate the design and implementation of vText linking video codecs with computer vision and image processing algorithms, and the performance evaluation shows that the vText framework achieves comparable running time and is easily used for prototyping visual  \nii  \nanalytical programs.  \nSecond, to reduce the vulnerability of deep neural networks against adversaries, we propose three color-reduction image processing approaches, which are Gaussian smoothing plus PNM color reduction (GPCR), Gaussian smoothing plus K-means (GK-means), and fast GK-means to make deep convolutional neural networks more robust to adversarial perturbation. We evaluate the approaches on a subset of the ImageNet dataset. Our evaluation reveals that our GK-means-based algorithms have the best top-1 classification accuracy.  \nThe final contribution of the dissertation is introducing a novel deep neural network compression framework on class specialization problems to address the limited utilization of deep neural network-based functionalities. We propose a novel knowledge distillation framework with two proposed losses, Renormalized Knowledge Distillation (RKD) and Intra-Class Variance (ICV), to render computationally efficient, specialized neural network models. Our quantitatively empirical evaluation demonstrates that our proposed framework achieves significant classification accuracy improvements for the tasks where the number of subcla","cbCain0e7qS2Q3Th","https://ap.wps.com/l/cbCain0e7qS2Q3Th","pdf",9412410,1,165,"English","en",105,"# Abstract\n# Introduction\n## Deployment of Machine Learning Systems\n## Security of Machine Learning Systems\n## Design of Machine Learning Systems\n## Summary\n# Video as Text: A New Paradigm for Flexible Video Analysis\n## Introduction\n## Related Work\n## Design of the vText Pipeline","[{\"question\":\"What motivates the dissertation’s focus on deployable and secure machine learning systems?\",\"answer\":\"Deep neural networks can be vulnerable to security attacks, and deploying such models is a major practical hurdle. The work targets both system design challenges and deployment difficulties while improving robustness.\"},{\"question\":\"How does the vText framework improve reusability for visual analytics?\",\"answer\":\"vText analyzes and manipulates video data similarly to how text is handled in Unix/Linux systems. It links video codecs with computer vision and image-processing algorithms to support flexible prototyping.\"},{\"question\":\"What techniques are proposed to improve adversarial robustness of deep neural networks?\",\"answer\":\"The dissertation proposes three color-reduction image-processing approaches: GPCR, GK-means, and fast GK-means. Experiments on a subset of ImageNet show the GK-means-based methods achieve the best top-1 classification accuracy.\"}]","Designing for Deployable, Secure, and Generic Machine Learning Systems - Dissertation Abstract | PDF",1785721934,416,{"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},"designing-for-deployable-secure-and-generic-machine-learning-systems-dissertation-abstract","",{"@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/designing-for-deployable-secure-and-generic-machine-learning-systems-dissertation-abstract/119018/",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","2026-08-03",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},"What motivates the dissertation’s focus on deployable and secure machine learning systems?","Question",{"text":75,"@type":76},"Deep neural networks can be vulnerable to security attacks, and deploying such models is a major practical hurdle. The work targets both system design challenges and deployment difficulties while improving robustness.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the vText framework improve reusability for visual analytics?",{"text":80,"@type":76},"vText analyzes and manipulates video data similarly to how text is handled in Unix/Linux systems. It links video codecs with computer vision and image-processing algorithms to support flexible prototyping.",{"name":82,"@type":73,"acceptedAnswer":83},"What techniques are proposed to improve adversarial robustness of deep neural networks?",{"text":84,"@type":76},"The dissertation proposes three color-reduction image-processing approaches: GPCR, GK-means, and fast GK-means. Experiments on a subset of ImageNet show the GK-means-based methods achieve the best top-1 classification accuracy.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]