{"649689":{"#nid":"649689","#data":{"type":"news","title":"Kwon Receives Honorable Mention for Dissertation at ISCA 2021","body":[{"value":"\u003Cp\u003E\u003Ca href=\u0022https:\/\/hyoukjunkwon.com\/\u0022\u003EHyoukjun Kwon,\u003C\/a\u003E a recent Georgia Tech Ph.D. graduate, received an Honorable Mention at the 2021 ACM-SIGARCH \/ IEEE-CS TCCA\u0026nbsp;Outstanding Dissertation Award ceremony, presented virtually\u0026nbsp;at the International Symposium on Computer Architecture (ISCA) 2021. ISCA 2021 was held June 14-19 and is the flagship venue for showcasing new ideas and research results in computer architecture.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EKwon\u0026rsquo;s award citation reads:\u0026nbsp;\u0026ldquo;For developing mechanisms to quantify the relationship between deep neural network mappings, data reuse, and communication flows for system design of flexible deep learning accelerators.\u0026rdquo; He is currently a research scientist at Facebook Reality Labs, where he has worked since October 2020. Kwon completed his Ph.D. under the guidance of \u003Ca href=\u0022https:\/\/tusharkrishna.ece.gatech.edu\/\u0022\u003ETushar Krishna\u003C\/a\u003E, who is the ON Semiconductor Junior Professor in the School of Electrical and Computer Engineering.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EMachine learning (ML), especially deep learning (DL), has demonstrated great results in computer vision, speech recognition, natural language processing, and recommendation systems. This has energized the entire field of computer architecture to develop customized hardware accelerators to enable deployment of DL solutions in the edge and cloud. Traditionally, the efficiency of domain-specific accelerators has come from\u0026nbsp;specialization, which takes place when the control path and datapath in the accelerator is tailored to the deep neural network (DNN). A key challenge facing the architecture community is that the field of ML is evolving very rapidly, so there is a very real possibility that silicon accelerator chips will be obsolete by the time they reach the market.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn order to solve this conundrum, Kwon\u0026rsquo;s\u0026nbsp;dissertation\u0026nbsp;proposes the key idea of\u0026nbsp;flexible\u0026nbsp;DNN accelerators. In contrast with a fully programmable processor, such as a CPU\/GPU, or a fully reconfigurable circuit, such as an FPGA, a flexible accelerator adds small-overhead-but-high impact reconfigurability for\u0026nbsp;future-proofing. Furthermore, Kwon\u0026rsquo;s thesis shows that these kinds of future-proofing\u0026nbsp;technologies can also improve performance on existing neural networks, as it allows the hardware to better tailor itself to the diverse set of layer parameters, instead of targeting the average case.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EKwon\u0026rsquo;s dissertation also develops a\u0026nbsp;foundational and formal understanding\u0026nbsp;of the complex interplay between the DNN model, its mapping, dataflow, memory accesses, communication flows, and microarchitectural choices. It also presents a suite of open-source software and hardware codebases demonstrating these ideas.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EKwon\u0026rsquo;s thesis has already shown measurable impact. Many foundational ideas are part of a\u0026nbsp;\u003Ca href=\u0022https:\/\/doi.org\/10.2200\/S01015ED1V01Y202005CAC052\u0022 target=\u0022_blank\u0022\u003ESynthesis Lectures\u003C\/a\u003E\u003Ca href=\u0022https:\/\/doi.org\/10.2200\/S01015ED1V01Y202005CAC052\u0022 target=\u0022_blank\u0022\u003E\u0026nbsp;on DNN accelerator design\u003C\/a\u003E, co-authored by Kwon; Michael Pellauer (NVIDIA), who mentored Kwon closely throughout his Ph.D. as a co-advisor; Angshuman Parashar (NVIDIA); Ananda Samajdar, a fellow Georgia Tech Ph.D. student; and Krishna. In addition, some of the open-source artifacts developed by the thesis, such as MAERI and\u0026nbsp;\u003Ca href=\u0022http:\/\/maestro.ece.gatech.edu\/\u0022 target=\u0022_blank\u0022\u003EMAESTRO\u003C\/a\u003E, are already being used by several research groups in industry, national labs, and other universities.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E###\u003C\/p\u003E\r\n\r\n\u003Cp\u003E* IEEE-CS TCCA stands for the IEEE Computer Society Technical Committee on Computer Architecture.\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EHyoukjun Kwon\u003Ca href=\u0022https:\/\/hyoukjunkwon.com\/\u0022\u003E,\u003C\/a\u003E a recent Georgia Tech Ph.D. graduate, received an Honorable Mention at the 2021 ACM-SIGARCH \/ IEEE-CS TCCA\u0026nbsp;Outstanding Dissertation Award ceremony, presented virtually\u0026nbsp;at the International Symposium on Computer Architecture (ISCA) 2021.\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"Hyoukjun Kwon, a recent Georgia Tech Ph.D. graduate, received an Honorable Mention at the 2021 ACM-SIGARCH \/ IEEE-CS TCCA Outstanding Dissertation Award ceremony, presented virtually at the International Symposium on Computer Architecture (ISCA) 2021."}],"uid":"27241","created_gmt":"2021-08-16 17:25:56","changed_gmt":"2021-08-16 17:31:39","author":"Jackie Nemeth","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2021-08-16T00:00:00-04:00","iso_date":"2021-08-16T00:00:00-04:00","tz":"America\/New_York"},"extras":[],"hg_media":{"649690":{"id":"649690","type":"image","title":"Hyoukjun Kwon","body":null,"created":"1629134866","gmt_created":"2021-08-16 17:27:46","changed":"1629134866","gmt_changed":"2021-08-16 17:27:46","alt":"photograph of Hyoukjun Kwon","file":{"fid":"246604","name":"hyoukjun.png","image_path":"\/sites\/default\/files\/images\/hyoukjun.png","image_full_path":"http:\/\/www.tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/hyoukjun.png","mime":"image\/png","size":1210239,"path_740":"http:\/\/www.tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/hyoukjun.png?itok=LlcLUbuO"}}},"media_ids":["649690"],"related_links":[{"url":"https:\/\/synergy.ece.gatech.edu","title":"Synergy Lab"},{"url":"http:\/\/www.ece.gatech.edu","title":"School of Electrical and Computer Engineering"},{"url":"http:\/\/www.gatech.edu","title":"Georgia Tech"},{"url":"https:\/\/iscaconf.org\/isca2021\/","title":"International Symposium on Computer Architecture 2021 (ISCA 2021)"}],"groups":[{"id":"1255","name":"School of Electrical and Computer Engineering"}],"categories":[{"id":"129","name":"Institute and Campus"},{"id":"130","name":"Alumni"},{"id":"134","name":"Student and Faculty"},{"id":"8862","name":"Student Research"},{"id":"135","name":"Research"},{"id":"153","name":"Computer Science\/Information Technology and Security"},{"id":"145","name":"Engineering"}],"keywords":[{"id":"183672","name":"Hyoukjun Kwon"},{"id":"173453","name":"Tushar Krishna"},{"id":"166855","name":"School of Electrical and Computer Engineering"},{"id":"179317","name":"Synergy Lab"},{"id":"109","name":"Georgia Tech"},{"id":"188652","name":"ACM-SIGARCH \/ IEEE-CS TCCA Outstanding Dissertation Award"},{"id":"180388","name":"International Symposium on Computer Architecture"},{"id":"12059","name":"IEEE Computer Society"},{"id":"172908","name":"Association for Computing Machinery"},{"id":"178072","name":"Deep Neural Networks"},{"id":"182279","name":"deep learning accelerators"},{"id":"9167","name":"machine learning"},{"id":"109581","name":"deep learning"},{"id":"11506","name":"computer vision"},{"id":"177001","name":"speech recognition"},{"id":"23981","name":"natural language processing"},{"id":"172677","name":"recommendation systems"},{"id":"175184","name":"computer architecture"},{"id":"181259","name":"hardware accelerators"},{"id":"177424","name":"edge"},{"id":"14704","name":"cloud"},{"id":"188653","name":"domain-specific accelerators"},{"id":"14549","name":"nvidia"},{"id":"180649","name":"MAERI"},{"id":"183681","name":"MAESTRO"}],"core_research_areas":[{"id":"39431","name":"Data Engineering and Science"},{"id":"39451","name":"Electronics and Nanotechnology"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003E\u003Ca href=\u0022mailto:jackie.nemeth@ece.gatech.edu\u0022\u003EJackie Nemeth\u003C\/a\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESchool of Electrical and Computer Engineering\u003C\/p\u003E\r\n\r\n\u003Cp\u003E404-894-2906\u003C\/p\u003E\r\n","format":"limited_html"}],"email":["jackie.nemeth@ece.gatech.edu"],"slides":[],"orientation":[],"userdata":""}}}