Updates
 [June 2021] Paper on Stronger Ensemble Learning of Probabilistic Circuits accepted at TPM Worshop, UAI'21
 [May 2021] Started an Applied Scientist Internship at Microsoft IDC
 [April 2021] Second runners up in shared task in NLP4IF workshop, NAACL'21. Paper accepted!
 [May 2020] Began remote research internship at StarAI Lab at UCLA
 [March 2020] Began remote internship at Computer Vision Lab at Linkoping Uniersity
 [May 2019] Selected for Google Summer of Code under The Julia Language
 [December 2018] Paper on Deep Learning for Traffic Sign Classification accepted at ICPRAM
 [July 2017] Started my journey at IIT Kharagpur as an undergraduate

Research
I am interested in Machine Learning in general, but currently focussing on Probabilistic Machine Learning and Bayesian Inference.
I aim to bring insights from a statistical and algorithmic viewpoint to build systems that work in the real world.
This amounts to building systems that can scale well to large dimensions and data while understanding when they are going wrong.
Tools from Probability Theory and Statistics can help answer such questions.
I am also interested developing robust deep learning algorithms for real worldapplications like autonomousdriving, robotics and healthcare.


Testing for a Change in Mean of a Weakly Stationary Time Series
Guide: Prof. Buddhananda Banerjee, IIT Kharagpur
Thesis /
Slides
Changepoint testing is the detection of an abrupt change in the distribution of a time series. Change in mean detection is generally posed as a hypothesis testing problem.
SelfNormalizing statistics aid such a detection by being adaptive to the presence and location of a changepoint.
We propose such a statistic that acheives sharper power rise compared to previous approahces on deviation from the nullhypothesis.


Independence Based Learning of Probabilistic Circuit Ensembles
Guide: Prof. Guy Van den Broeck, Yitao Liang, StarAI Lab, UCLA
Paper /
Slides /
Poster /
Code /
TPM Workshop, UAI'21
StructuredDecomposable Probabilistic Circuits model the data by encoding different types of contextspecificindependences (CSIs).
One can use these CSIs as prior information for initializing the mixture components of an EMensemble of circuits.
We propose an algorithm by partitioning the data in coherence with these CSIs and training a circuit on each of them to get a stronger initialization of an ensemble.


Graph Representation Learning for Road Networks
Guide: Prof. Michael Felsberg, Zahra Gharaee, Linkoping University
Paper /
Pattern Recognition Journal
Real world road networks can be represented as a graph with intersections as nodes and roads as edges. One can use graphbased deep learning approaches to learn efficient representations of such nodes.
We survey various algorithms in the literature for this task and propose a DFS based aggregation scheme to aggregate information from nodes far away from a given node.


Predicting Binary Properties of Tweets : NAACL Shared Task
Paper /
Code /
Task Link /
NLP4IF Workshop, NAACL 2021
Tweets during the time of a pandemic, if contain false information, can lead to spread of misinformation. The task was thus to automatically answer multiple binary questions (in yes/no) pertaining to a tweet.
This problem was framed as a multioutput learning problem with answering of each question formulated as a task.
Intertask attention modules were proposed for aggregating information on top of a largelanguagemodel.
Approach resulted in runnersup position.


Perception for Autonomous Driving
Guide: Prof. Debashish Chakravarty, Autonomous Ground Vehicle Research Group, IIT Kharagpur
Report /
Video /
Code
Perception systems are crucial for robots to make sense of their surroundings.
Developed an endtoend objectdetection and tracking pipeline for trafficsign detection and classification for Indian Signs.
Built a realtime road segmentation newtork for roadregion segmentation. Both modules tested on electric vehicle Mahindrae2o.


Uncertainty Estimation in Deep Learning
Course: Advanced Machine Learning, Prof. Pabitra Mitra
Report
Surveyed multiple papers on calibration and uncertainty in deep neural networks. Methods like MCDropout, Bayes by Backprop, SGLD and pSGLD were implemented and tested on MNIST.
Uncertainty estimates of the predictions were obtained on outofsample images as a proof of concept.


Machine Learning for Portfolio Optimization
Course: Optimization Methods in Finance, Prof. Geetanjali Panda
Report
Built forecasting models to predict future stock prices of 30 stocks using LSTMs, Linear Regression and Support Vector Regression.
Forecasts were used to select top `N` performing stocks to build a portfolio. Markowitz models were applied for portfolio optimization.
Use of forecasts lead to better sharperatios and returns in general on the testset.


Regression Analysis for Medical Cost Estimation
Course: Regression and Time Series, Prof. Buddhananda Banerjee
Report
Built statistical linear regression models to predict medical costs of patients.
Improved Rsquared value of the fit by implementing and analysing residualsvsfitted, sclaelocation, residualsvsleverage and normalqq plots.


Google Summer of Code
Blog
Implemented multimodal image2image translation GANs like pix2pix, cycleGAN SRGAN purely in The Julia Language.
Wrote a library for reinforcementlearning containing PPO and TRPO implementations.
Implemented NeuralImageCaptioning in a pure julia codebase.


128bit AESFeistel Encryption
Course : Cryptography and Network Security, Guide : Prof. Sourav Mukopadhyay
Code
Implemented a hybrid block cipher for encrypting inputs of length 128bits. The first 5 rounds are AES followed by 5 rounds of a custom Feistel Cipher with variable number of blocks. Finally 5 rounds of AES follow.


Autonomous Navigation Robot for IGVC'18&19
Intelligent Ground Vehicle Competition Entry from IIT Kharagpur, Prof. Debashish Chakravarty
Report /
Video
Built the planning and perception stack for the Autonomous Navigation Challenge at Intelligent Ground Vehicle Competition'19, Michigan University, USA.
The robot had to follow a set of GPS waypoints autonomously, remaining within lane boundaries and avoiding obstacles.
Used SVMs for object detection and superpixel clustering for lane segmentation.
Entry bagged runnersup position and won the earlyqualification round.

Of course I did not design this page, template adopted from here

