Talks and Presentations
Below is a list of selected oral/poster presentations, conference participation, and research activities.
Oral and Poster Presentations
Adaptive Conformal Semantic Uncertainty in LLMs for Responsible AI
Oral and Poster Presentations at Trust to Impact: Trustworthy and Responsible Al for Citizenship, Migration, and Social Good (TIAI) Workshop
Localized Uncertainty Quantification in Medical Image Segmentation via Adaptive Conformal Prediction
Poster presentation at the Vector Institute Research Symposium
Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction
Oral presentation at COPA 2024
Quantifying Deep Learning Model Uncertainty in Conformal Prediction
Oral presentation at AAAI Human AI 2023
Evidence-based Subjective Classification Cost in Deep Neural Networks
Poster presentation at the Vector Institute Research Symposium
Subjective Uncertainty in Federated Learning
Poster presentation at the Vector Institute Research Symposium
Ontology Alignment Using Inductive Logic Programming
Oral presentation at ICWR
Participation and Activities
The 13th Symposium on Conformal and Probabilistic Prediction with Applications (COPA)
UN PET Lab: Hackathon Competition
Participated in a privacy-preserving learning hackathon focused on safe information sharing and training models under differential privacy constraints (epsilon privacy budget). The setting involved unknown training labels with multiple queries and evaluation on an unseen test set with a score based on accuracy, parameter values, and privacy budget spent.