Journal ·
MK-SGC-SC: Multiple Kernel guided Sparse Graph Construction in Spectral Clustering for Unsupervised Speaker Diarization
Nikhil Raghav, Avisek Gupta, Swagatam Das, Md Sahidullah
arXiv
Deep Learning · AI/ML · TCG CREST
Learning the Hidden Complexity of the World
The Representations Learning Lab develops the theoretical foundations and algorithms for artificial intelligence that can grasp why events occur, moving beyond mere pattern recognition. Our core focus spans causality, data representation, and advanced machine learning.
4+
Publications
2
Researchers
2
Research Areas
2024
Founded
Research Areas
View all →Featured Publications
All papers →Lab members in bold.
Conference ·
Kernelized Low-rank matrix recovery: Application in collaborative representation based classification
Subhajit Saha, Angshul Majumdar, Swagatam Das, Avisek Gupta
International Joint Conference on Neural Networks (IJCNN)
Paper ↗
DOI:10.1109/IJCNN64981.2025.11227597
low-rank matrix recovery nuclear norm minimization kernel methods Majorisation Minimization Collaborative Representation based Classification face recognition
Journal ·
From Wasserstein to Maximum Mean Discrepancy Barycenters: A Novel Framework for Uncertainty Propagation in Model-Free Reinforcement Learning
Srinjoy Roy, Subhajit Saha, Swagatam Das
IEEE Transactions on Emerging Topics in Computational Intelligence
The Team
Meet everyone →We are a diverse group united by a passion for the mathematical foundations of learning.
Avisek Gupta
Principal Investigator
Subhajit Saha
PhD Student