{"619103":{"#nid":"619103","#data":{"type":"event","title":"PhD Defense by Divya Mahajan","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003ETitle:\u0026nbsp;\u003C\/strong\u003EBalancing Generality and Specialization for Machine Learning in the Post ISA Era\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDivya Mahajan\u003C\/p\u003E\r\n\r\n\u003Cp\u003EPhD Candidate\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESchool of Computer Science\u003C\/p\u003E\r\n\r\n\u003Cp\u003ECollege of Computing\u003C\/p\u003E\r\n\r\n\u003Cp\u003EGeorgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E------------------------\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EDate: \u003C\/strong\u003EFriday, March 15, 2019\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ETime: \u003C\/strong\u003ENoon - 2:00 PM\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ELocation: \u003C\/strong\u003EKlaus 2100\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E------------------------\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ECommittee:\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Hadi Esmaeilzadeh (\u003Cem\u003EAdvisor\u003C\/em\u003E), Department of\u0026nbsp;Computer Science and Engineering, University of California, San Diego\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Hyesoon Kim, School of Computer Science, Georgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Milos Prvulovic, School of Computer Science, Georgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Doug Burger, Microsoft Corporation\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Dean Tullsen, Department of Computer Science and Engineering, University of California, San Diego\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E------------------------\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EA growing number of commercial and enterprise systems are increasingly relying on compute-intensive machine learning algorithms.\u0026nbsp;While the demand for these applications is growing, the performance benefits from general-purpose platforms is diminishing.\u0026nbsp;This challenge has coincided with the explosion of data where the rate of data generation has reached an overwhelming level that is beyond the capabilities of current computing systems.\u0026nbsp;Therefore, the ever-increasing compute needs of applications such as machine learning and robotics can\u0026nbsp;benefit from hardware acceleration.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003ETraditionally, to accelerate a set of workloads, we profile the code optimized for CPUs and offload the hot functions on compute units designed specially for that\u0026nbsp;particular function, hence providing higher performance and energy efficiency.\u0026nbsp;Instead in this work, we take a revolutionary approach where we delve into the algorithmic properties of an application domain\u0026nbsp;and couple them with our hardware acceleration solutions. We leverage the property that a wide range of\u0026nbsp;machine learning algorithms can be modeled as\u0026nbsp;stochastic optimization problems; and use this property\u0026nbsp;to devise comprehensive stacks\u0026nbsp;that are built\u0026nbsp;independent of the CPU. These stacks\u0026nbsp;expose a high-level mathematical programming interface and can automatically generate accelerators for users who have limited knowledge about hardware design but can benefit from large performance and efficiency gains for their programs.\u0026nbsp;\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EKeeping these ambitious goals in mind, our work (1) strikes a balance between generality and specialization by breaking the long-held traditional abstraction of the Instruction Set Architecture (ISA) in favor of a more algorithm-centric approach;\u0026nbsp;(2) develops hardware acceleration frameworks by co-designing a language, compiler, runtime system, and hardware to provide high performance and efficiency, in addition to flexibility and programmability;\u0026nbsp;(3) segregates algorithmic specification from implementation to shield the programmer from continual hardware\/software modifications while allowing them to benefit from the emerging heterogeneity of modern compute platforms;\u0026nbsp;and (4) develops real cross-stack prototypes to evaluate these innovative solutions in a real-world setting and make them open-source to maximize community engagement and industry impact.\u0026nbsp;Our work Tabla (\u003Ca href=\u0022http:\/\/act-lab.org\/artifacts\/tabla\/\u0022\u003Ehttp:\/\/act-lab.org\/artifacts\/tabla\/\u003C\/a\u003E)\u0026nbsp;is public, and defines the very first open-source hardware platforms for machine learning and artificial intelligence.\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":": Balancing Generality and Specialization for Machine Learning in the Post ISA Era"}],"uid":"27707","created_gmt":"2019-03-11 18:04:48","changed_gmt":"2019-03-11 18:04:48","author":"Tatianna Richardson","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2019-03-15T13:00:00-04:00","event_time_end":"2019-03-15T15:00:00-04:00","event_time_end_last":"2019-03-15T15:00:00-04:00","gmt_time_start":"2019-03-15 17:00:00","gmt_time_end":"2019-03-15 19:00:00","gmt_time_end_last":"2019-03-15 19:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"100811","name":"Phd Defense"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"78771","name":"Public"},{"id":"174045","name":"Graduate students"},{"id":"78751","name":"Undergraduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}