Below are some publicly disclose-able projects I have worked on in the past.
Feel free to contact me if you are interested in collaborating on a project or research direction.
Probabilistically Check-able Proofs (PCP) (Class Project) attempts to highlight the power of PCP’s and demonstrates their use case in several applications domains. Additionally we describe the weak PCP theorem in detail. [Report]
Generalization Bound for Metric and Similarity Learning (Class Project) attempts to estimate generalization bound for metric and similarity learning using Rademacher Complexity and general matrix norms. (2022)
DARPA Explainable Artificial Intelligence (XAI) attempts to explain artificial intelligence model decisions, funded by the Defense Advanced Research Projects Agency (DARPA). (2019 – 2021)
Iterative and Adaptive Sampling with Spatial Attention for Black-Box Model Explanations. [Paper] [Conference Video][Oral video] [Slides] [Poster]
Jersey number reader with dual head joint learning. Learn decimal representation with neural networks. (2018) [Code]
Visualizing Resiliency in Neural Network Interpretations for satellite imagery. (2018) (Master’s thesis) [Thesis][Paper1][Paper2][Paper3][Journal Paper]
