Technical Talks

[April 2025] Multi-label Recognition under Noisy Supervision: A Confusion Mixture Modeling Approach, ICASSP 2025, Hyderabad, India

[April 2025] Tutorial: End-to-End Learning from Crowdsourced Labels: A Signal Processing Perspective, ICASSP 2025, Hyderabad, India

[Oct 2024] Deep Learning under Instance-dependent Label Noise:A Tensor Factorization Perspective, Asilomar 2024, Pacific Grove, CA

[Oct 2024] Robust Learning for Artificial Intelligence, AI Club Seminar, NIT Calicut, India

[Mar 2024] Ensuring Robustness in Machine Learning by Combating Real-world Data Uncertainties, CECS Seminar, University of Central Florida, Orlando, FL

[Aug 2023] Provably Robust Learning: A Tale of Tackling Label Noise through Naïve Bayes to Deep Neural Networks, Invited Talk, Washington State University, Pullman, WA

[Jun 2023] Towards Efficient Learning under Label Noise: From Dawid-Skene to Deep Neural Networks, Invited Talk, University of Central Florida, Orlando, FL

[Jun 2023] Under-Counted Tensor Completion with Neural Side Information Learner: Recoverability and Algorithm, SIAM OP23, Seattle, WA

[May 2023] Deep Learning From Crowdsourced Labels: Coupled Cross-Entropy Minimization, Identifiability, and Regularization, ICLR Virtual Talk

[Mar 2023] Learning from Noisy Labels with Theoretical Guarantees, Invited Talk, University of Texas, Arlington, TX

[July 2021] Crowdsourcing via Annotator Co-occurrence Imputation & Provable Symmetric Nonnegative Matrix Factorization, ICML Virtual Talk

[Jun 2021] Learning Mixed Membership from Adjacency Graph via Systematic Edge Query: Identifiability and Algorithm, ICASSP Virtual Talk

[Nov 2020] Recovering Joint PMF from Pairwise Marginals, Asilomar Signal Processing Conference Virtual Talk

[Jun 2019] Stochastic Optimization for Coupled Tensor Decomposition with Applications in Statistical Learning, IEEE Data Science Workshop, Minnesota, MN

[Mar 2019] Crowdsourcing via Pairwise Co-occurrences: Identifiability & Algorithms, Artificial Intelligence Seminar, Oregon State University

[Feb 2019] Crowdsourcing via Pairwise Co-occurrences: Identifiability & Algorithms, Signal Processing Seminar, Oregon State University