Anupriya Kumari
Core Member, DSG IIT Roorkee
Education
- B.Tech, Electronics and Communication Engineering, IIT Roorkee (2022–2026)
Research Interests
- Trustworthy AI
- Explainable AI
- Computer Vision
Publications
-
Singhi, S., Kumari, A. Strengthening Interpretability: An Investigative Study of Integrated Gradient Methods._ TMLR 2025.
-
Bhardwaj, D., Kumari, A., Jindal, S. Rethinking Randomized Smoothing from the Perspective of Scalability._ NeurIPS 2024 Workshop.
Projects
- Deepfake Face Detection
Reproducibility study of the Haar Wavelet Transform-based deepfake detection method, proposed a number of corrections, including proper edge structure classification, structural analysis validated through Laplacian sharpness measurements, and precise kernel specifications. - Open Source
Contributed to SciPy (PRs #19666) as a part of SciPy Open Source Sprint conducted by Pyladies, Berlin. - Multi-Approach Image Classification
Investigated the results of the K-means clustering algorithm, transfer learning techniques, and scratch-trained convolutional neural network architectures to achieve Image classification using Tensorflow and Keras.
Follow
| GitHub |