Jaedong Hwang

About

I am a master's student at Computer Vision Lab (advisor: Bohyung Han), Seoul National University (SNU). From June 2020 to March 2021, I did an internship at Adobe Research under the guidance of Joon-Young Lee and Seoung Wug Oh. During undergraduate study, I worked as a research intern at the lab and T-Brain. I also have spent great time with Joseph Lim at the University of Southern California and Byoung-Tak Zhang at BioIntelligence Lab, SNU.

My research interests come from a question, "How humans can learn from small amounts of data or coarse data such as noisy or weak-labeled data and why machines cannot?" My goal is to reduce the gap between human and machine intelligence by deeply understanding human intelligence and to build robust machines for these coarse data insipired by cognitive science. Shortly, I am interested in these fields as follows:

• Computer Vision: video understanding, weakly-/self-supervised learning, open-set/long-tailed recognition
• Reinforcement Learning: embodied AI, dynamics randomization
• Psychology: curiosity, intuitive concepts (e.g., intuitive physics), decision-making, categorization, child development
• Neuroscience: visual pathway, procedural memory