santhan-sukhija

Sanatan Sukhija

Assistant Professor
sanatan.sukhija@mahindrauniversity.edu.in

Dr. Sanatan Sukhija is currently working as an Assistant Professor in the Department of Computer Science and Engineering at Mahindra University, Hyderabad. His varied career includes stints at several industries and academic institutions, including, Amazon, Intel, Siemens, HCL, Punjab Engineering College, Chandigarh and the NorthCap University, Gurugram.

He has worked on a few fundamental Machine Learning and Deep Learning problems that focus on generalizability through transfer learning and domain adaptation, inspired by application domains such as Activity Recognition, Computer Vision, Cross-lingual Analysis, and Influence Maximization in Social Networks. His other research deals with learning robust deep models for industry/healthcare related problems. Specifically, his research focuses on learning in those domains where the amount of labeled training data is scarce. His research has led to multiple publications at several top-tier venues (AI Journal, AAAI, IJCAI, ECML-PKDD, IJCNN, WCCI etc.).

He is looking for Ph.D. students who would like to work on interesting and challenging inter-disciplinary problems in other fields, including but not limited to, ubiquitous computing, and pervasive computing.

 

2014-2020

Ph.D.
Indian Institute of Technology Ropar
July 2014 – Jan 2020
Thesis Title: Leveraging Label Space Similarities for Transfer Learning
Thesis Advisor: Narayanan C Krishnan http://cse.iitrpr.ac.in/ckn/people/ckn.html

2011-2013

M. Tech.
Computer Engineering
Malaviya National Institute of Technology Jaipur
July 2011 – May 2013

2007-2011

B. Tech.
Computer Science and Engineering
Maharaja Agrasen Institute Of Technology, GGSIPU, New Delhi
Aug 2007 – May 2011

Academia:

  • Punjab Engineering College, Chandigarh
    Assistant Professor, 14/10/19 to 17/07/20
  • The NorthCap University (formerly ITM), Gurugram, Haryana
    Assistant Professor, 1/7/13 to 21/6/14

Industry:

  • Intel Technology India Private Limited, Bangalore
    Graduate Research Intern, 6/8/18 to 4/1/19
  • Amazon, Bangalore
    Applied Scientist Intern, 4/6/18 to 4/8/18
  • HCL, Noida
    SDE Intern, 1/7/10 to 3/8/10
  • NetTech, BITS Goa
    Trainee (Network Administration on Linux Servers), May 2010 to June 2010
  • Siemens, Gurgaon
    SDE Intern, 1/8/09 to 30/9/09

Journals

2019

  • Sanatan Sukhija and Narayanan C Krishnan, “Supervised Heterogeneous Feature Transfer via Random Forests,” Artificial Intelligence Journal, Elsevier, Volume 268, March 2019, Pages 30-53.

    • Book Chapters

    2020

  • Sanatan Sukhija and Narayanan Chatapuram Krishnan, “Shallow Domain Adaptation”, Domain adaptation in computer vision with deep learning, (editors) Hemanth Venkateswara and Sethuraman Panchanathan, Springer, 2020.

    • International Conferences

    2021

  • Somyadeep Shrivastava, Dheeraj Chaudhary, Yayati Gupta, and Sanatan Sukhija. "Cost Effective Influence Maximisation." In 2021 International Conference on COMmunication Systems & NETworkS (COMSNETS), pp. 90-93. IEEE, 2021.

    • 2020

  • Sanatan Sukhija, Srenivas Varadarajan, Narayanan C Krishnan and Sujit Rai “Multi-Partition Feature Alignment Network for Unsupervised Domain Adaptation,” IJCNN, 2020.

    • 2018

  • Sanatan Sukhija and Narayanan C Krishnan, “Web-Induced Heterogeneous Transfer Learning with Sample Selection,” in Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 2018, pp. 777-793.
  • Sanatan Sukhija, “Label Space Driven Heterogeneous Transfer Learning with Web Induced Alignment,” in Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018.
  • Sanatan Sukhija, “Label Space Driven Feature Space Remapping,” in Proceedings of the ACM India Joint International Conference on Data Science and Management of Data, 2018, pp. 310-313. [Best Paper Presenter Award]
  • Sanatan Sukhija, Narayanan C Krishnan and Deepak Kumar, “Supervised Heterogeneous Transfer Learning via Random Forests,” in Proceedings of the ACM India Joint International Conference on Data Science and Management of Data, 2018, pp. 157-166.

    • 2017

  • Monika Dahiya, Sanatan Sukhija, Deepak Kumar and Hukum Singh, “Image Encryption based on watermarking technique in Fourier domain using Pixel scrambling,” in Proceedings of the Fourth IEEE International Conference on Image Information Processing, 2017.

    • 2016

  • Sanatan Sukhija, Narayanan C Krishnan and Gurkanwal Singh, “Supervised Heterogeneous Domain Adaptation via Random Forests,” in Proceedings of the 25th International Joint Conference on Artificial Intelligence, 2016, pp. 2039-2045.

    • 2014

  • Monika Dahiya, Sanatan Sukhija and Hukum Singh, "Image encryption using quad phase masks in fractional Fourier domain and case study," in Proceedings of IEEE international advance computing conference, 2014.

    • 2013

  • Sanatan Sukhija, Subhash Panwar and Neeta Nain, “CRAMM: character recognition aided by mathematical morphology,” in Proceedings of the IEEE Second International Conference on Image Information Processing, 2013.

    • Areas of Interest: Machine Learning (Theoretical and Applied), Transfer Learning and Deep Learning.

    About my Research:

    Given plentiful labeled training data, deep learning really shines when it comes to complex problems such as image classification, natural language processing, and speech recognition. I have worked on specific machine learning problems, in particular, transfer learning and domain adaptation. This research area focuses on learning in those domains where the amount of labeled training data is scarce or absent. Preferably, for the next few years, I intend to extend my research to the deep transfer context where the ideas from shallow transfer algorithms can be leveraged to learn robust deep models for industry/healthcare related problems or applications inspired from multimedia and IoT. I am also open to working on interesting and challenging inter-disciplinary problems in other fields, including but not limited to, data mining, ubiquitous computing, and pervasive computing. Auto-ML, Adversarial Learning, Meta-Learning, Unsupervised Learning, Zero-Shot Learning, and Reinforcement Learning are some of the other areas that interest me. I am looking forward to working with students who are technically strong in Mathematics and Computer Science.

    • Professional Affiliation
      Member:ACM, AAAI and IEEE.
    • Professional Service
      Reviewer: IEEE TKDE, International Journal of Intelligent Systems (Wiley), IEEE ICCV, WACV, and ICIP.