I was a recent student spotlight for ML@GT
Projects
*
Continual Diffusion: Continual Customization of Text-to-Image Diffusion with C-LoRA
We learn to generate text-conditioned images of new concepts in a sequential manner (i.e., continual learning) without forgetting.
CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learning
We learn decomposed prompts for rehearsal-free continual learning.
ConStruct-VL: Data-Free Continual Structured VL Concepts Learning
We continuously adapt BLIP to new VL concepts.
Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning
Data-Free Class-Incremental Learning, where an agent must learn new concepts over time without storing generators or training data from past tasks
Memory-Efficient Semi-Supervised Continual Learning: The World is its Own Replay Buffer
Realistic Semi-Supervised Continual Learning, where data distributions reflect object class correlations between, and among, the labeled and unlabeled data distributions
Unsupervised Progressive Learning and the STAM Architecture
Learning data representations from a non-stationary stream of unlabeled data for downstream classification tasks