Publications

. Big but Invisible Adversarial Attacks. In submission, 2018.

. Detecting Anomalous Faces without Labeled Examples. In submission, Also accepted to CVPR-W(VBSD), Poster at Midwest Computer Vision Workshop, 2018.

Poster

. Improving Style Transfer with Calibrated Metrics. In submission, Poster at Midwest Computer Vision Workshop, 2018.

Preprint Poster

Projects

Active Learning for GCN Based NER

Graph Convolutional Network(GCN) in Active Learning using Uncertainty Sampling for Named Entity Recognition. [code] [report] [youtube]

Advanced Image Generation Methods

Lecture video and a survey paper on Advanced Image Generation Methods.

ML based Predictions of Indian Rail Tickets

ML models to predict if (and when) a wait-listed ticket will get confirmed [code] [report]

Teaching Experience

University of Illinois

  • CS498 Applied Machine Learning (Fall 2018)
  • CS225: Data Structures (Spring 2017)
  • CS101: Intro to computer science (Spring 2016* & Fall 2017)
    *(Excellent TA)

Talks

  • [June’18] Presented our work “Detecting Anomalous Faces with ‘No Peeking’ Autoencoders” at CVPR’s VBSD Workshop [slides]
  • [April’18] My slides on ConceptGAN and Mix & match networks from CVPR’18 presented at UIUC’s Vision Lunch [.ppt, .pdf]
  • [Mar’18] Presented two posters at the Midwest Computer Vision Workshop on Detecting Anomalous Faces and Quantitative Evaluation of Style Transfer
  • [Dec’17] My presentation slides on Deformable Convolution Networks from ICCV 2017 at UIUC’s Vision Lunch
  • [Mar’17] Our lecture video on Advanced Image Generation Methods can be found here on youtube

Contact