I am a PhD Candidate at Virginia Tech (Fall 2019 - present). I work under the supervision of Dr. Hoda Eldardiry and Dr. Ismini Lourentzou. I received my Master of Science from Virginia Tech in Spring 2022. I completed my BS in Computer Science and Engineering from Bangladesh University of Engineering and Technology in March 2019.
Email: afrina@vt.edu
My research interests are Machine Learning, particularly designing novel representation learning objectives for multi-modal data. My ongoing projects aim to address the limitations of current machine learning models under scarce and/or high annotation cost scenarios. Current research directions include (1) improving the efficiency of contrastive self-supervised methods for learning robust representations, (2) addressing the temporal coherence challenges in multi-video summarization, (3) extending my work on zero-shot video generation
[2023.12] Our paper uaMix-MAE: Efficient Tuning of Pretrained Audio Transformers with Unsupervised Audio Mixtures (internship work @ Microsoft Research) got accepted at ICASSP'2024!!
[2023.9] Our PlanBot team has advanced to AlexaPrize TaskBot Challenge 2 finals!
[2023.7] Our PlanBot team has advanced to the AlexaPrize TaskBot Challenge 2 semifinals!
[2023.5] I’m thrilled to start my Research internship at Microsoft Research this summer!
[2023.4] I’m thrilled to pass my Preliminary Examination!
[2022.5] I’m thrilled to start my internship at Los Alamos National Laboratory (LANL) this summer!
