Hi, I'm Pratyaksh Gautam, a student researcher at IIIT Hyderabad!
My research is primarily concerned with how deep neural networks process and organize information for auditory tasks. I’m also interested in interpretability at large, specifically for control-based applications (model editing, unlearning, and continual learning).
I’m working my Master’s thesis, advised by Prof. Vinoo Alluri and Prof. Makarand Tapaswi. My work on my thesis is on concept-based interpretability of deep neural networks for music, specifically looking at the causal relationships between lower-level concepts like instruments and higher-level concepts like genre.
Part of my research on this was done as a research assistant at the University of Jyväskylä (JYU), under the supervision of Prof. Petri Toivainen. Previously, I've done internships at Wells Fargo, working on semantic search in a resource-constrained environment, and Texas Instruments, working on single-stage scene text spotting methods.