I’m a Ph.D. candidate at Georgia Tech advised by Prof. Faramarz Fekri. My research primarily focuses on Machine Learning, Probabilistic graphical models and Causal Discovery. In particular, I’m interested in understanding the feasibility of recovering graph structure from observational and interventional data. Previously, I got my Masters from Georgia Tech and I did my undergraduate studies at National Institute of Technology, Trichy (NITT).
In my free time I like to do a bit of cooking and explore hiking trails around Atlanta.
news
Sep 28, 2023
Presented my work on Density evolution based sensing matrix design for covariance recovery at Allerton’23.
Apr 27, 2023
Presented my work on cyclic causal graph discovery (NODAGS-Flow) at AISTATS’23.
May 16, 2022
I interned at Genentech over the summer (2022) working on Causal discovery.
Dec 14, 2021
Presented my work on “Visual Question Answering based on Formal logic” at ICMLA, 2021.
Learning causal relationships between variables is a well-studied problem in statistics, with many important applications in science. However, modeling real-world systems remain challenging, as most existing algorithms assume that the underlying causal graph is acyclic. While this is a convenient framework for developing theoretical developments about causal reasoning and inference, the underlying modeling assumption is likely to be violated in real systems, because feedback loops are common (e.g., in biological systems). Although a few methods search for cyclic causal models, they usually rely on some form of linearity, which is also limiting, or lack a clear underlying probabilistic model. In this work, we propose a novel framework for learning nonlinear cyclic causal graphical models from interventional data, called NODAGS-Flow. We perform inference via direct likelihood optimization, employing techniques from residual normalizing flows for likelihood estimation. Through synthetic experiments and an application to single-cell high-content perturbation screening data, we show significant performance improvements with our approach compared to state-of-the-art methods with respect to structure recovery and predictive performance.
A Density Evolution Framework for Recovery of Covariance and Causal Graphs from Compressed Measurements
Sethuraman, Muralikrishnna G., Zhang, Hang, and Fekri, Faramarz
In Fifty Ninth Annual Allerton Conference on Communication, Control, and Computing 2023
In this paper, we propose a general framework for designing sensing matrix A, for estimation of sparse covariance matrix from compressed measurements of the form y = Ax + n. By viewing covariance recovery as inference over factor graphs via message passing algorithm, ideas from coding theory, such as Density evolution (DE), are leveraged to construct a framework for the design of the sensing matrix. The proposed framework can handle both (1) regular sensing, i.e., equal importance is given to all entries of the covariance, and (2) preferential sensing, i.e., higher importance is given to a part of the covariance matrix. Through experiments, we show that the sensing matrix designed via density evolution can match the state-of-the-art for covariance recovery in the regular sensing paradigm and attain improved performance in the preferential sensing regime. Additionally, we study the feasibility of causal graph structure recovery using the estimated covariance matrix obtained from the compressed measurements.
Visual Question Answering based on Formal Logic
Sethuraman, Muralikrishnna G., Payani, Ali, Fekri, Faramarz, and Kerce, J. Clayton
In 2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA) 2021
Visual question answering (VQA) has been gaining a lot of traction in the machine learning community in the recent years due to the challenges posed in understanding information coming from multiple modalities (i.e., images, language). In VQA, a series of questions are posed based on a set of images and the task at hand is to arrive at the answer. To achieve this, we take a symbolic reasoning based approach using the framework of formal logic. The image and the questions are converted into symbolic representations on which explicit reasoning is performed. We propose a formal logic framework where (i) images are converted to logical background facts with the help of scene graphs, (ii) the questions are translated to first-order predicate logic clauses using a transformer based deep learning model, and (iii) perform satisfiability checks, by using the background knowledge and the grounding of predicate clauses, to obtain the answer. Our proposed method is highly interpretable and each step in the pipeline can be easily analyzed by a human. We validate our approach on the CLEVR and the GQA dataset. We achieve near perfect accuracy of 99.6% on the CLEVR dataset comparable to the state of art models, showcasing that formal logic is a viable tool to tackle visual question answering. Our model is also data efficient, achieving 99.1% accuracy on CLEVR dataset when trained on just 10% of the training data