Girish Varma
Assistant Professor Center for Security, Theory & Algorithms Research (CSTAR) Machine Learning Lab (MLL) Kohli Center on Intelligent Systems (KCIS) IIIT HyderabadHi!
I am an Assistant Professor at the Center for Security, Theory & Algorithms Research (CSTAR) and the Machine Learning Lab at IIIT Hyderabad, broadly interested in theoretical and applied computer science problems.
I did my Phd thesis on Computational Complexity theory (at TIFR Mumbai) and continues to have interest in Complexity theory, Combinatorics and Graph theory. Later I went on to do research (at IIIT Hyderabad) in Deep Learning, specifically Model Compression techniques and Semantic Segmentation for Autonomous Navigation. A summary of my research can be found here.
Teaching
Modern Complexity Theory (Monsoon 2020)
A first undergraduate course on Computational Complexity Theory.
Probability & Statistics (Monsoon 2020)
A first course in probability and statistics with a focus on discrete spaces.
Probabilistic Graphical Models
Probabilistic Graphical Models refers to i.) concise representations of probability distributions using graphs ii.) efficient algorithms for sampling distributions represented in such form iii.) learning these representations from data.
Advanced Mathematical Structures
This course is intended to familiarize the students with the types of mathematical reasoning found in theoretical research on computing and communications. The course contains a broad set of intermediate and advanced level topics in Algebra, Combinatorics, Probability and Graph Theory.
Machine Learning in Natural Sciences
A project oriented course in applying machine learning techniques to solving problems in natural sciences.
Efficient CNNs
Surveying methods used to make deep learning models efficient.
Events & Talks

IDD Challenge at NCVPRIPG’19
Open for Submisions till 17th Dec, 2019. 
Scene Understanding Challenge for Auto. Nav. in Unstructured Environments
Second International Workshop On Auto. Nav. in Unconstrained Environments
in Conjunction with ICCV 2019, November 2, 2019, Seoul. 
Counting Objects under Symmetry: Pólya  Burnside Enumeration

Theory Seminars at IIIT Hyderabad. June 1, 2019

Semantic Segmentation: Recent Advances, Indian Datasets & Universal Segmentation Problems
Challenges and Advances in VisionBased SelfDriving. Tutorial cohosted with Prof. Manmohan Chandrakar (UCSD)
at ICVGIP’18, Hyderabad. December, 2018. 
IDD Dataset and Challenge
ICTAI Workshop on Unstructured Driving Scenarios in India
Intel Bangalore, November, 2018. 
Scene Understanding Challenge for Auto. Nav. in Unstructured Environments
First International Workshop On Auto. Nav. in Unconstrained Environments in Conjunction with ECCV 2018, September 8, 2018, Munich. 
3rd Summer School on Machine Learning  Computer Vision
IIIT Hyderabad, July, 2018. 
Autorickshaw Detection Challlenge at NCVPRIPG’17
Tutorial on Computer Vision for Autonomous Navigation at NCVPRIPG’17
IIT Mandi, December, 2017. 
2nd Summer School on Machine Learning  Computer Vision
IIIT Hyderabad, July, 2017.
Group
Research Areas
Tags
Selected Publications
Ramanujan Bipartite Graph Products for Efficient Block Sparse Neural Networks
Dharma Teja Vooturi, Girish Varma, Kishore Kothapalli
arXiv
Deep Learning Enabled Inorganic Material Generator
Yashaswi Pathak, Karandeep Singh Juneja, Girish Varma, Masahiro Ehara, U. Deva Priyakumar
ChemArXiv
Using Artificial Intelligence (AI) to Classify Retinal Developmental Disorders
Helen Kuht, Shuihua Wang, Garima Nishad, Sharon George, Gail Maconachie, Viral Sheth, Zhanhan Tu, Michael Hisaund, Rebecca McLean, Seema Teli, Frank Proudlock, Girish Varma, YuDong Zhang, Irene Gottlob, Mervyn George Thomas
Investigative Ophthalmology & Visual Science
Semantic Segmentation Datasets for Resource Constrained Training
Ashutosh Mishra*, Sudhir Kumar*, Tarun Kalluri*, Girish Varma, Anbumani Subramanian, Manmohan Chandrakar, C V Jawahar
7th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG 2019)
Dynamic Block Sparse Reparametarization of Convolutional Neural Networks
Dharma Teja Vooturi, Girish Varma, Kishore Kothapalli
4th International Workshop on
Compact and Efficient Feature Representation and Learning in Computer Vision, ICCV 2019
Universal Semisupervised Semantic Segmentation
Tarun Kalluri, Girish Varma, Manmohan Chandraker, C V Jawahar
International Conference on Computer Vision
IDD: A Dataset for exploring problems in Autonomous Navigation in Unconstrained Environments
Girish Varma, Anbumani Subramanian, Anoop Namboodiri, Manmohan Chandraker, C V Jawahar
IEEE Winter Conference on Applications of Computer Vision
Deep Expander Networks: Efficient Deep Networks from Graph Theory
Ameya Prabhu, Girish Varma, Anoop Namboodiri
Oral Presentation
European Conference on Computer Vision
Improved Visual Relocalization by Discovering Anchor Points
Soham Saha, Girish Varma, C V Jawahar
Spotlight Presentation
British Machine Vision Conference
Cityscale Road Audit System using Deep Learning
Sudhir Kumar Reddy, Girish Varma, C V Jawahar
JTCF Novel Technology paper award Finalist
International Conference on Intelligent Robots
Hardness of Approximate Coloring
Phd Thesis advised by Prof. Prahlad Harsha
Supported by Google India Phd Fellowship in Algorithms
Tata Institute of Fundamental Research, Mumbai.
A Characterization of Hardtocover CSPs
Amey Bhangale, Prahladh Harsha, Girish Varma
Theory of Computing Journal (ToC)
Computational Complexity Conference (CCC)
Superpolylogarithmic hypergraph coloring hardness via lowdegree long codes
Venkat Guruswami, Prahladh Harsha, Johan Hastad, Srikanth Srinivasan, Girish Varma
SIAM Journal on Computing (SICOMP)
Sym. of Theory of Computing (STOC)
Physarum Can Compute Shortest Paths
Vincenzo Bonifaci, Kurt Mehlhorn, Girish Varma
Journal of Theoretical Biology
Symp. of Discrete Algorithms
Contact
Email: firstname [dot] lastname [at] iiit.ac.in
Phone: 1212 (prefix for external +91406653)
Address: Rm: A3118, Vindya A2, Center for Security, Theory and Algorithms Research, IIIT Hyderabad