Abstract: Click here
About the speaker: Dr. Swarnendu Sil is an assistant professor at IISc Bangalore, specializing in geometric analysis, partial differential equations and calculus of variations. After transitioning from a background in engineering at Jadavpur University to a Ph. D. at EPFL, he held postdoctoral positions at both EPFL and ETH Zürich. More about him: https://math.iisc.ac.in//~ssil/
Abstract: Increasingly large and complex spatial datasets pose massive inferential challenges due to high computational and storage costs. Our study is motivated by the KAUST Competition on Large Spatial Datasets 2023, which tasked participants with estimating spatial covariance-related parameters and predicting values at testing sites, along with uncertainty estimates. We compared various statistical and deep learning approaches through cross-validation and ultimately selected the Vecchia approximation technique for model fitting. To overcome the constraints in the R package GpGp, which lacked support for fitting zero-mean Gaussian processes and direct uncertainty estimation, two features necessary for the competition, we developed additional R functions. Additionally, we implemented subsampling-based approximations and parametric smoothing for estimators with skewed sampling distributions. Our team, DesiBoys, comprised of Rishikesh Yadav, currently an Assistant Professor at IIT Mandi, Pratik Nag, currently a Postdoctoral Fellow at the University of Wollongong, and I from IIT Kanpur, secured first place in two of the four sub-competitions and second place in the other two, validating the effectiveness of our proposed strategies. Moreover, we extended our evaluation to a large, real-world spatial satellite-derived dataset of total precipitable water, comparing the predictive performance of different models using multiple diagnostics. If time permits, we will discuss additional experiences with various data challenge competitions that we successfully participated in over the last few years.
About the speaker: Dr. Arnab Hazra is an assistant professor at IIT Kanpur. He received his Ph. D. in 2018 from North Carolina State University, Raleigh. His research interests are too numerous to list in such a short introduction; please see his webpage for more details: https://sites.google.com/view/arnabhazra09/home
Abstract: Round Surgery Diagrams for 3-manifolds.
In this talk, we introduce round surgery diagrams in S^3 as a natural analogue of Dehn surgery diagrams for constructing 3-manifolds. We establish a precise correspondence between a natural class of round surgery diagrams and Dehn surgery diagrams in S^3. Consequently, every closed connected oriented 3-manifold can be obtained by round surgery on a framed link in S^3, recovering Asimov’s result. Different round surgery presentations can yield the same 3-manifold. We define four local moves on round surgery diagrams and prove that any two diagrams presenting the same 3-manifold are related by a finite sequence of these moves, yielding a Kirby Calculus for round surgery. As an application, we show that 3-manifolds obtained by round 1-surgery on two-component fibred links in S3 admit taut foliations, hence carry tight contact structures. This is a joint work with Dr. Prerak Deep and part of his doctoral thesis.
About the speaker: Dr. Dheeraj Kulkarni is an assistant professor of mathematics at IISER Bhopal. He received his Ph. D. in 2012 from IISc Bengaluru under the supervision of Professor Siddhartha Gadgil. His research interests are in Geometry and Topology, and in particular Contact and Symplectic Topology.
More about him: https://sites.google.com/iiserb.ac.in/dheerajkulkarni
Abstract: The routes to chaos and the global bifurcations leading to chaotic behavior are two fascinating areas of research in nonlinear dynamics.
Chaotic dynamics are observed in a wide range of mathematical models across various disciplines of science and engineering. In recent years, the structural sensitivity of models with respect to their bifurcation structures leading to chaos has received increasing attention. The main objective of this talk is to discuss the structural sensitivity of the bifurcation structure associated with the classical Hastings–Powell model and the global bifurcations that give rise to chaotic regimes in the modified Lorenz system. A systematic bifurcation analysis, incorporating both local and global bifurcations, provides deeper insights into the routes to chaos and the nature of transient dynamics. The techniques discussed here can also be applied to problems arising in other areas of science and technology.
About the speaker: Dr. Malay Banerjee is a Professor in the Department of Mathematics and Statistics at IIT Kanpur, where he has served since April 2008. His research interests include: Mathematical Ecology, Nonlinear Dynamics, Mathematical Epidemiology, Spatio-temporal Pattern Formation. Dr. Banerjee earned his Ph.D. in Applied Mathematics from the University of Calcutta in 2005.
Abstract: Life is full of complex, evolving systems — from markets to environmental systems. Using tools from mathematics, statistics, physics, and AI, scientists can unravel patterns hidden within large datasets. This talk would offer a glimpse into how we decode real-world complexity using data science.
About the speaker: Dr. Anirban Chakraborti is a Professor at Jawaharlal Nehru University and a Fellow of the World Academy of Sciences (FTWAS). He is a founding member of the Centre for Complexity Economics, Applied Spirituality and Public Policy at O. P. Jindal Global University, and an International Member of the Centro Internacional de Ciencias AC. His work focuses on complexity science, econophysics, and computational approaches to social and economic systems.
More about him and his group: http://www.jnu.ac.in/Faculty/anirban/index.html