Manik Varma (computer scientist)

Manik Varma is an Indian computer scientist and a senior principal researcher at Microsoft Research India.[1] He also holds an adjunct professor of computer science position at the Indian Institute of Technology Delhi. He was a Rhodes Scholar and earned his PhD from the University of Oxford under the guidance of Andrew Zisserman working on Texture Classification in Computer Vision. He also held a post-doctoral fellowship at Mathematical Sciences Research Institute, Berkeley before joining Microsoft Research.

Not to be confused with Manik Varma, an Indian classical singer.
Manik Varma
NationalityIndian
Alma materUniversity of Oxford
AwardsShanti Swarup Bhatnagar Prize for Science and Technology (2019)
Scientific career
FieldsComputer Science

Computer Vision
Machine Learning

Computational Advertising
InstitutionsMicrosoft Research India

Indian Institute of Technology Delhi
MSRI

University of California, Berkeley
ThesisStatistical approaches to texture classification (2004)
Doctoral advisorAndrew Zisserman
Websitehttp://manikvarma.org/

He currently conducts research in the broad fields on Machine Learning, Computer Vision and Computational Advertising. In 2013, he started and popularized a new area in machine learning called Extreme Classification (also known as Extreme Multi-label Classification).[1][2][3] Extreme Classification focuses on Multi-Label Classification at the scale of millions of labels and helps rethink traditional problems of ranking and recommendation.[4] Extreme Classification is thriving in both academia and industry with product integrations in Bing and Amazon.[4][1] Manik Varma along with his colleagues at MSR India also proposed another paradigm in machine learning called Edge Machine Learning[5] to enable machine learning predictions on tiny IoT devices with as little as 2 KB of RAM assisting in low-energy, low-latency and privacy preserving applications of AI. In the past, he worked on statistical approaches to texture classification, object detection, multiple kernel learning and ranking.

He has been awarded the prestigious Shanti Swarup Bhatnagar Prize for Science and Technology, one of the highest Indian science awards for his contributions to Engineering Sciences in 2019.[6][7] His research works won the WSDM Best Paper award[8] and BuildSys Best Paper Runner-Up award[9] in 2019. He is also an Elected Fellow of Indian National Academy of Engineering[10] and held the Visiting Miller Professorship at University of California, Berkeley.[11][1] His professional services include being an associate editor for the IEEE Transactions on Pattern Analysis and Machine Intelligence.[12]

References
  1. "Manik Varma". manikvarma.org. Retrieved 2020-07-26.
  2. "The Extreme Classification Repository". manikvarma.org. Retrieved 2020-07-26.
  3. Bengio, Samy; Dembczynski, Krzysztof; Joachims, Thorsten; Kloft, Marius; Varma, Manik (2019). Bengio, Samy; Dembczynski, Krzysztof; Joachims, Thorsten; Kloft, Marius; Varma, Manik (eds.). "Extreme Classification (Dagstuhl Seminar 18291)". Dagstuhl Reports. 8 (7): 62–80. doi:10.4230/DagRep.8.7.62. ISSN 2192-5283.
  4. Varma, Manik. "Extreme Classification". cacm.acm.org. Retrieved 2020-07-26.
  5. microsoft/EdgeML, Microsoft, 2020-07-25, retrieved 2020-07-26
  6. "Awardee Details: Shanti Swarup Bhatnagar Prize". ssbprize.gov.in. Retrieved 2020-07-26.
  7. Bureau, Our (27 September 2019). "Microsoft researcher Manik Varma among Shanti Swaroop Bhatnagar award winners this year". @businessline. The Hindu.
  8. "Home | 12th ACM International WSDM Conference". www.wsdm-conference.org. Retrieved 2020-07-26.
  9. "BuildSys 2019". buildsys.acm.org. Retrieved 2020-07-26.
  10. "Nomination Information". Indian National Academy of Engineering. Retrieved 2020-07-26.
  11. "2018 - 2019 Lunch Lectures". miller.berkeley.edu. Retrieved 2020-07-26.
  12. "TPAMI Editorial Board | IEEE Computer Society". Retrieved 2020-07-26.
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.