← All Research Tracks

Interpretability & Theory of DL

Worked on developing attribution techniques using Integrated Gradients, Manifold IG, and Guided IG towards neural network interpretability.

  • Currently extending these methods to sequential models and large language models (LLMs) using Granger causality and Randomized Path Integrals.
  • Reproduced the results of Integrated Gradients and Manifold Integrated Gradients and tested a new methodology on MIG to improve performance.
  • Future work involves computer vision applications.

Explainable AI: Attribution Techniques

Theory of LLMs — Notion Notes

Calendly - Let's meet!