Publications

📝 f-INE: Influence Estimation using Hypothesis Testing

Published in ICML, 2025 workshop on DataWorld, 2025

🎯 The principal objective of this work is to propose a hypothesis testing framework for estimating the influnce of training data for a model prediction.

Recommended citation: https://openreview.net/pdf?id=26zj5O7dph

📝 Variational Diffusion Unlearning: a variational inference framework for unlearning in diffusion models

Published in NeurIPS, 2024 Workshop on SafeGenAi, 2024

🎯 The principal objective of this work is to propose a machine unlearning methodology that can prevent the generation of outputs containing undesired features from a pre-trained diffusion model.

Recommended citation: https://openreview.net/pdf?id=B2wDjiED9V

📝 Concept Forgetting via Label Annealing

Published in 41st Conference on Uncertainty in Artifical Intelligence (UAI,2025), 2024

🎯 The primary aim of this study is to propose a mathematical definition of forgetting in ML models and develop methodologies for forgetting specific undesired concepts from pre-trained classification models.

Recommended citation: https://openreview.net/pdf?id=2L7KQ4qbHi

📝 Partially Blinded Unlearning: Class Unlearning for Deep Networks a Bayesian Perspective

Published in The 39th Annual AAAI Conference on Artificial Intelligence (AAAI,2025), 2024

🎯 The principal aim of this study is to formulate a methodology tailored for the purposeful elimination of information linked to a specific class of data from a pre-trained classification network.

📝 FAST: Feature Aware Similarity Thresholding for Weak Unlearning in Black-Box Generative Models

Published in IEEE Transactions on Artificial Intelligence (TAI,2024), 2023

🎯 The primary goal of this study is twofold: first, to elucidate the relationship between filtering and unlearning processes, and second, to formulate a methodology aimed at mitigating the display of undesirable outputs generated from models characterized as black-box systems.

📝 Adapt then Unlearn: Exploiting Parameter Space Semantics for Unlearning in Generative Adversarial Networks

Published in Transactions on Machine Learning Research (TMLR,2025), 2023

🎯 The objective of this work is to prevent the generation of outputs containing undesired features from a pre-trained GAN where the underlying training data set is inaccessible.

Recommended citation: https://arxiv.org/abs/2309.14054

📝 Driving a Charged Coupled Device (CCD) by microcontroller for LIBS based application

Published in IEEE International Symposium on Devices Circuits and Systems (ISDCS), 2018

🎯 The objective of this project is to utilize a LASER INDUCED BREAKDOWN SPECTROGRAPHY (LIBS) instrument for soil mineral composition analysis in agriculture, facilitated by an Android-controlled platform.

Subhodip Panda