Gautam Dasarathy

Honored to receive the 2024 Top 5% Teaching Award from ASU's Fulton Schools of Engineering. Grateful for the recognition and motivated to continue enhancing student educational experiences.

Participated as a panelist in 'Implementing Generative AI in Your Courses', hosted by FSE Learning and Teaching Hub in collaboration with IEEE Educational Activities.
[YouTube]

We have a new paper, 'Active Sequential Two-Sample Testing', accepted to the Transactions on Machine Learning Research. Author list: Weizhi Li, Prad Kadambi, Pouria Saidi, Karthikeyan Natesan Ramamurthy, Gautam Dasarathy, Visar Berisha.
[Open Review]

Two exciting preprints went live:

• 'On Characterizations of Potential and Ordinal Potential Games' by S Arefizadeh, A Nedich, G Dasarathy. [arXiv]
• 'Unraveling Overoptimism and Publication Bias in ML-driven Science' by P Saidi, G Dasarathy, V Berisha. [arXiv]

Delivered an invited talk at Informs IOS 2024 in Houston, TX on 'Efficiently Solving a System of Complex Quadratic Equations: Optimization Landscape and Statistics'. This is based on joint work with Parth Thaker and Angelia Nedic

Parth Thaker successfully defended his excellent thesis that explored a range of areas including bandits, interactive learning, multi-agent systems, and nonconvex optimization. After impressive internships at MERL and Intuitive Surgical, Parth is joining Intuitive Surgical full-time.

New paper in Wiley Stat on 'Nonparanormal Graph Quilting with Applications to Calcium Imaging', co-authored by A Chang, L Zheng, G Dasarathy, and G Allen.
[link]

The following paper was accepted to the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2023

• A. Rayas, R. Anguluri, J. Cheng, G. Dasarathy Differential Analysis of Networked Systems that Obey Conservation Laws.

Two papers from our group have been accepted at the Neural Information Processing Systems (NeurIPS) 2022 conference:

• A. Rayas, R. Anguluri, G. Dasarathy Learning the Structure of Large Networked Systems that Obey Conservation Laws.
• P. Thaker, M. Malu, N. Rao, G. Dasarathy. Maximinzing and Satisficing in Multi-armed Bandits with Graph Information.

Very excited to be a co-PI on a fantastic multi-disciplinary project as part of the NSF PIPP program. This is led by Pavan Turaga

A multi-disciplinary project has been accepted for support as part of the DARPA Geometries of Learning (GoL) program. This is led by Pavan Turaga and the full team includes Visar Berisha (ASU), Vishal Patel (JHU), Rama Chellappa (JHU), Anuj Srivastava (FSU).

Our paper titled "A Multisensory Approach to Present Phonemes as Language Through a Wearable Haptic Device" was declared the winner of the 2022 IEEE Transaction on Haptics Best Application Paper Award!

I am extremely honored to receive the Distinguished Alumni Award (Academics) from my alma mater VIT University, Vellore.

Our paper on phylogenetic inference was accepted to the Journal of Mathematical Biology (JoMB).

It is an absolute honor to be recognized for my contributions to ASU's teaching mission. All credit goes to the fantastic students who make teaching stimulating!

A new project titled "Active Meta Learning" has been accepted for support by Office of Naval Research (ONR). This project will push the frontiers of meta-features in data-starved regimes and is led by Visar Berisha.

I am honored to have received the NSF Faculty Early Career Development (CAREER) Award for a research program that aims to develop novel theory and algorithms for learning and leveraging the structure of graphs. Thanks NSF!
[Read More]

A paper titled "Graph Community Detection from Coarse Measurements: Recovery Conditions for the Coarsened Weighted Stochastic Block Model" has been accepted to the 2021 International Conference on AI & Statistics (AISTATS) in the oral track (top ~3% of 1500+ submissions).

This is joint work with Nafiseh Ghoroghchian and Stark Draper from the University of Toronto.

A paper on active learning algorithms for estimating topological properties of decision boundaries accepted has been accepted to NeurIPS 2020.

This is joint work with Weizhi Li, Visar Berisha from ASU and Karthikeyan Ramamurthy from IBM.
[arXiv] [proceedings]

A new project titled "Distributed Learning over Multi-Access Channels: From Bandlimited Coordinate Descent to Gradient Sketching" has been funded by the National Science Foundation.

This is a collaboration with Junshan Zhang (ASU), Na Li (Harvard), Zhi Ding (UC Davis). Thanks NSF!

A new project titled "Alpha Loss: A New Framework for Understanding and Trading Off Computation, Accuracy, and Robustness in Machine Learning " has been funded by the National Science Foundation.

This is a collaboration with Lalitha Sankar (ASU). Thanks NSF!
[Read More]

A new project titled "Graphical Models from Partially Observed Interactions with Biomedical Applications" has been funded by the National Institutes of Health.

This is a collaboration with Genevera Allen (Rice University). Thanks NIH!

A paper titled "Differentiable Programming for Hyperspectral Unmixing using a Physics-based Dispersion Model" has been accepted to the 2020 European Conference on Computer Vision.

This is joint work with John Janiczek, Parth Thaker, Christopher Edwards, Phil Christensen, Suren Jayasuriya

A new project titled "Active Tracking of Disease Spread in COVID19 via Graph Predictive Analytics" has been funded by the National Science Foundation under the RAPID program.

This is a cross-disciplinary collaboration at ASU with Doug Cochran (Math), Huan Liu (CS), Patricia Solis (Geography), and Pavan Turaga (AME). Thanks NSF!
[Read More] [Local News Coverage]

Two paper have been accepted to the IEEE International Symposum on Information Theory:

• Sypherd, T., Diaz, M., Sankar, L., and Dasarathy, G. On the alpha-loss Landscape in the Logistic Model.
• Thaker, P., Dasarathy, G., and Nedić, A. On the Sample Complexity and Optimization Landscape for Quadratic Feasibility Problems.

I am honored to have received the Top 5% Teaching Award from the Fulton Schools of Engineering.
[Read More]