Physics CS ML

resume, github

Peer-Reviewed Publications

• A. Hasan, I. Rugina, A. Wang; Pruning for Protection: Increasing Jailbreak Resistance in Aligned LLMs Without Fine-Tuning; BlackBoxNLP @ EMNLP 2024
• I. Rugina, R. Dangovski, L. Jing, P. Nakov, M. Soljačić; Data-Informed Global Sparseness in Attention Mechanisms for Deep Neural Networks; LREC-COLING 2024
• I. Rugina, R. Dangovski, M. Veillette, P. Khorrami, B. Cheung, O. Simek, M. Soljačić; Meta-Learning and Self-Supervised Pretraining for Storm Event Imagery Translation; IEEE HPEC 2023
• P. Khorrami, O. Simek, B. Cheung, M. Veillette, R. Dangovski, I. Rugina, M. Soljačić, P. Agrawal; Adapting Deep Learning Models to New Meteorological Contexts Using Transfer Learning; IEEE Big Data 2021

Class Projects

• Last Iterate Convergence of the EG Method for Variationally Coherent Min-Max Problems: Write-up, Slides
• Meta-learning for Multi-Task Symbolic Regression: Write-up, Slides, Code
• Physics-Inspired Phase-Coded Bayesian Neural Networks: Write-up

Notes

These notes contain direct excerpt from relevant papers/ lecture notes/ videos.

• Vector and Matrix norms
• Hopfield networks
• Non-Equilibrium stat mech
• Stat-mech Inspired Methods
• VAE
• Median of Means
• Diffusion Models
• GPU Puzzles
• Optimal Brain Compression