[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124096-en":3,"doc-seo-124096-105":30,"detail-sidebar-cat-0-en-105":92},{"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":20,"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},124096,687197207639,"Asher","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Machine learning applied to GPU test generation - Master’s thesis","Investigates how to generate efficient stimuli for simulation-based verification of digital circuits, where verification often depends on constrained random testing and human effort, making the process tedious, slow, and costly. Proposes and implements a complete optimization system for random-constrained verification using Bayesian optimization to tune testbench weights and adds a scaling mechanism to extend controllable weights. Evaluates the approach on two devices, measuring absolute coverage and compute costs. Results show performance comparable to random verification, with gains when combined with human input and slight advantages over constrained random, while limitations remain for cases without additional input or under the proposed scaling.","Master’s thesis  \nNT NU  \nNorwegian Un iversity of Science and Technology  \nFaculty of Information Techno logy and Electrical Engineering Department of Electronic Systems  \nJonas Tørnes  \nMachine learning applied to GPU test generation  \nMaster’s thesis in Electronics Systems Design and Innovation Supervisor: Bjørn B. Larsen  \nCo-supervisor: Øystein Gjermundnes June 2023  \nJonas Tørnes  \nMachine learning applied to GPU test generation  \nMaster’s thesis in Electronics Systems Design and Innovation Supervisor: Bjørn B. Larsen  \nCo-supervisor: Øystein Gjermundnes June 2023  \nNorwegian University of Science and Technology  \nFaculty of Information Technology and Electrical Engineering Department of Electronic Systems  \nPreface  \nThis master’s thesis has been written as the final assignment for the five-year master’s program in Electronics Systems Design and Innovation at the Norwegian University of Science and Technology. It is submitted as the final delivery in the course ”TTFE4940 1 Electronics Systems Design and Innovation, Master’s thesis”. It has been supervised by Prof. Bjørn B. Larsen from the Department of Electronic Systems and DEng. Øystein Gjermundnes at Arm Norway AS. All the work in this thesis has been carried out during the spring semester of 2023 . It is a spiritual continuation of the specialization project conducted during the autumn of 2022 [1], where the problem was the same, but the solution proposed was different.  \nAcknowledgements  \nThe work with this thesis has had its ups and downs, frustrating and rewarding simultaneously. But also, never have I learned so much from a project; it has been exciting to take a deep dive into machine learning and try and use it for something useful and new. Spending the last year thinking about this problem has been a privilege, and it is with a bit of sadness that the journey is finally over. I would first like to express my gratitude to my supervisors throughout this last year. Thankyou, Bjørn, for our bi-weekly meetings and for listening and allowing me to share my thoughts and frustrations. An Øystein, you have constantly been encouraging and positive through this project, even when facing new problems or things did not go according to the plan. You have also always pushed me to write a better thesis; your input has been invaluable. I would also like to express my gratitude to Wade Walker for creating the idea which resulted in this thesis. Zhirong Yang from the Department of Computer Science has also been an enormous resource for anything machine learning related and helped by proposing the initial solution.  \nFive years is quite a long time, or at least I thought it would be that, but it has passed too quickly. That would not have been the case had it not been for all the amazing people I have met through these five years here in Trondheim; I am very grateful for that. Finally, I want to thank my family, Mom, Dad, and my sister Anne for supporting me through these five years and the 18 years prior. I could not have done this without you.  \nAbstract  \nThis thesis has investigated how to produce and find efficient stimuli for simulation-based verification of digital circuits. The current state-of-the-art verification still relies heavily on random constrained verification and human involvement in the verification process. This results in tedious work or the need to simulate many tests to ensure good design coverage; this process is timeconsuming and costly. Currently, very few methods exist, allowing this process to be optimized with machine learning.  \nThis thesis proposes and implements a complete system for optimizing random-constrained verification. It utilizes Bayesian optimization to optimize the weights assigned inside the testbench. Because of the current limitations of Bayesian optimization, we have also implemented a way to scale the maximum number of weights the system can control. The system is tested on two devices to verify its capabilities. Both absolute co","cbCaie4R7sXOq3mQ","https://ap.wps.com/l/cbCaie4R7sXOq3mQ","pdf",14917033,1,78,"English","en",105,"# Abstract\n## Motivation and problem statement\n## Proposed system and methods\n## Evaluation setup and metrics\n## Results and key findings","[{\"question\":\"What problem does the thesis address in digital circuit verification?\",\"answer\":\"It targets the inefficiency of simulation-based verification that relies on constrained random testing and human involvement, which can be time-consuming and costly while still requiring many simulations for adequate coverage.\"},{\"question\":\"How does the proposed system optimize test generation?\",\"answer\":\"It uses Bayesian optimization to tune the weights inside the testbench, aiming to produce efficient stimuli and improve verification effectiveness.\"},{\"question\":\"What were the main results of the method on two test devices?\",\"answer\":\"The method generally performed about as well as random verification, but it could improve over random when used with human input and showed slightly better behavior than constrained random in some scenarios, with device-specific improvements.\"}]","Machine learning applied to GPU test generation - Master’s thesis | PDF",1785820295,197,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-applied-to-gpu-test-generation-masters-thesis","",{"@graph":36,"@context":86},[37,54,69],{"@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/machine-learning-applied-to-gpu-test-generation-masters-thesis/124096/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-04",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the thesis address in digital circuit verification?","Question",{"text":76,"@type":77},"It targets the inefficiency of simulation-based verification that relies on constrained random testing and human involvement, which can be time-consuming and costly while still requiring many simulations for adequate coverage.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the proposed system optimize test generation?",{"text":81,"@type":77},"It uses Bayesian optimization to tune the weights inside the testbench, aiming to produce efficient stimuli and improve verification effectiveness.",{"name":83,"@type":74,"acceptedAnswer":84},"What were the main results of the method on two test devices?",{"text":85,"@type":77},"The method generally performed about as well as random verification, but it could improve over random when used with human input and showed slightly better behavior than constrained random in some scenarios, with device-specific improvements.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]