Gianmarco Jhair Gallardo

About me

PhD candidate at RIT working on self-supervised learning, continual learning, and vision foundation models.

Hello! I'm a PhD candidate in Imaging Science at Rochester Institute of Technology (RIT), advised by Dr. Christopher Kanan. My research focuses on self-supervised learning, continual learning, and vision foundation models. I'm particularly interested in learning transferable visual representations and developing models that can adapt efficiently to new tasks and changing data distributions.

During my PhD, I've worked on developing efficient continual learning methods, including memory replay and sleep-inspired learning approaches. I've also investigated out-of-distribution generalization and transfer learning in pretrained vision models. My research has led to publications at ICML, NeurIPS, TMLR, and BMVC.

I also have industry research experience in medical imaging AI. At Siemens Healthineers, I worked on self-supervised pretraining of 3D vision models for medical imaging, investigating how learned representations can improve training efficiency for downstream tasks such as lung nodule detection, segmentation, and classification.

I received my BS in Mechatronics Engineering from Universidad Nacional de Ingeniería in Peru, where my work included deep learning for diabetic retinopathy detection and SLAM for mobile robot exploration.

You can find my papers on the Publications page and my full CV on its own page.

News

Show older newsHide older news