Avatar

Prashant Shekhar

Postdoctoral fellow

Tufts University

Biography

Hi, I’m Prashant. Currently I am a postdoc at Tufts University’s Data Intensive Studies Center (DISC), working with Dr. Abani Patra . Currently, my work mainly focusses on learning hierarchical sparse representations for large datasets enabling faster inference. I am interested in using machine learning and physics to understand evolution of large cryospheric systems like polar icesheets.

Interests

  • Machine Learning and AI
  • Hierarchical sparse models
  • Large scale bayesian models
  • Glaciology
  • Remote sensing

Education

  • PhD in Computational Data Science, 2019

    University at Buffalo

  • MSc in Mechanical Engineering, 2016

    University at Buffalo

  • BTech in Industrial Engineering, 2014

    Indian Institute of Technology, Kharagpur

Experience

 
 
 
 
 

Postdoctoral fellow

Tufts University

Oct 2019 – Present 419 Boston Ave, Medford, MA 02155
 
 
 
 
 

Research/Teaching Assistant

University at Buffalo

Aug 2014 – Aug 2019 Buffalo, NY 14260
 
 
 
 
 

Research Assistant

Indian Institute of Technology, Kharagpur

May 2014 – Jul 2014 Kharagpur, West Bengal 721302, India
 
 
 
 
 

Student Assistant

The Hong Kong Polytechnic University

May 2013 – Jul 2013 11號 Yuk Choi Rd, Hung Hom, Hong Kong

Projects

Hierarchical Machine Learning

Exploring mathematics behind sparse hierarchical modeling approaches

Studying Polar Ice sheets

Using Machine learning to understand ice sheet evolution .

Publications

Shekhar, P., Csatho, B., Schenk, T., Roberts, C., Patra, A. (2019). ALPS: A framework for modeling time series of land ice changes. To be submitted to IEEE Transactions on Geoscience and Remote Sensing.

Project

Shekhar, P., Patra, A. (2019). Hierarchical regularization networks for sparsification based learning on noisy datasets.. To be submitted to SIAM Journal on Mathematics of Data Science (SIMODS).

Project

Shekhar, P., Patra, A. (2019). Hierarchical Data Reduction and Learning. To be submitted to SIAM Journal on Mathematics of Data Science (SIMODS).

Project

Shekhar, P., Patra, A., Csatho, B. (2017). Multiscale and Multiresolution methods for Sparse representation of Large datasets.. Procedia Computer Science 108: 1652-1661.

Project

Shekhar, P., Patra, A., Stefanescu, ER. (2016). Multilevel methods for sparse representation of topographical data.. Procedia Computer Science 80: 887-896.

Project

See all publications

Honors

Dean’s graduate fellowship

University at Buffalo Sep 2014

Best Undergraduate thesis award

Indian Institute of Technology, Kharagpur Apr 2014

Contact

  • 419 Boston Avenue, Medford, MA, 02155, United States