BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium: Beyond ISAs: Striking a Balance between Generality 
 and Specialization for AI/ML
DTSTART:20220321T090000
DTEND:20220321T100000
DTSTAMP:20260506T205401Z
UID:9d78a873e438bbc3485a801901d13a750ccb429d70d56989f9a24b58
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Divya Mahajan - Microsoft\nIC Faculty candidate\n\nAbstrac
 t\nAdvances in Artificial Intelligence (AI) and Machine Learning (ML) are 
 beginning to revolutionize medicine\, manufacturing\, commerce\, transport
 ation\, and other key aspects of our lives. However\, such transformative 
 effects are predicated on providing high-performance compute capabilities 
 to enable these learning algorithms. Domain specific accelerators are an e
 fficient and performant means to meet the compute requirements of these la
 rge-scale AI/ML. As the new age data-centers become heterogeneous with the
 se emerging domain specific hardware\, we must rethink both the architectu
 re and the corresponding system stack.\nIn this talk\, I will provide an o
 verview of my contributions to design\, deploy\, and utilize accelerators 
 for a wide class of AI/ML applications. I will first discuss pioneering wo
 rks TABLA and DaNA\, which are comprehensive full-stack solutions for mach
 ine learning accelerators that integrate with data management systems. The
 se solutions expose a high-level programming interface to programmers that
  have limited knowledge about hardware design\, nevertheless\, can benefit
  from performance and efficiency gains through acceleration. Then\, I will
  describe FAE\, a novel framework that leverages statistical properties of
  data to best utilize the heterogeneous compute and memory resources for r
 ecommender model training. Finally\, I will conclude with my future resear
 ch vision towards devising architectures and systems for sustainable massi
 ve-scale distributed AI/ML by exploring the challenges which arise from th
 e cross-pollination of different components in the data processing pipelin
 e.\n\nBio\nDivya Mahajan is a Senior Researcher in the Cloud Accelerated S
 ystems & Technologies group at Microsoft. She leads the research\, design\
 , and deployment of communication primitives for massive-scale distributed
  deep learning. She obtained her PhD in Computer Science from Georgia Inst
 itute of Technology. She obtained her Masters from The University of Texas
  Austin\, Texas and Bachelors from Indian Institute of Technology Ropar. H
 er research interests lie in designing novel architectures and building ro
 bust systems to address the needs of new and emerging applications. She is
  passionate about continuing innovative research to have a broad impact on
  computing and society in general.\nDivya is the recipient of National Cou
 ncil for Women and Information Technology Collegiate Award\, President of 
 India Gold Medal at IIT\, and has been a Finalist in the Qualcomm Innovati
 on Fellowships. Her work has been recognized with the College of Computing
  Dissertation Award\, HPCA Distinguished Paper Award\, and has appeared in
  top architecture\, database\, systems\, and machine learning venues like 
 ISCA\, MICRO\, HPCA\, ASPLOS\, VLDB\, NeurIPS and high impact journals lik
 e IEEE Micro.\n\nMore information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420 https://epfl.zoom.us/
 j/63078472550?pwd=QXYrQ1JCZ3lsRFRYa0J0U1JDSlNVQT09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR