I am a Ph.D. student at Rensselaer Polytechnic Institute (RPI). My research applies the fundamentals of Energy-based modeling and Dense Associative Memory (also known as modern Hopfield networks) for the designing of new neural architectures and understanding the interplays between memorization and generalization in deep generative models, such as diffusion models.
During my time in grad school, I am fortunate to work with Dmitry Krotov, a wonderful physicist working at IBM Research (where I also spent my summers at), who happens to be the co-inventor of Dense Associative Memory alongside John J. Hopfield. Our works have been highlighted by IBM Research (see articles 1 and 2) and in this article from Quanta Magazine.
To learn more about my works, please refer to my blog for further details.
