Thesis on studying the problem of specifying privacy requirements for machine learning based systems, in a manner that combines interpretability with high utility by specifying privacy requirements as constraints, thereby allowing for both interpretability and automated optimization of the utility.

Contributions include:

• Study of privacy requirements for machine learning based systems, in a manner that combines interpretability with high utility
• Specification of privacy requirements as constraints, thereby allowing for both interpretability and automated optimization of the utility
• Incorporating privacy requirements in a language model to generate privacy preserving ML systems
• Developing privacy preserving ML-LLM systems for multi-centric studies

Research Interest:

• Privacy-Preserving Machine Learning
• Large Language Models
• Privacy-Preserving Generative AI
• Natural Language Processing
• Statistical Machine Learning

Coursework:

• Artificial Intelligence
• Machine Learning
• Human Computer Interaction
• Software Engineering
• Probabilistic Graphical Models and their Applications
• Neural Networks: Implementation and Application
• The Elements of Statistical Learning
• Statistical Natural Language Processing
• Statistics with R
• Image Acquisition Methods
• Text to Speech Synthesis
• Seminar: Non-Supervised Machine Learning Techniques
• Project Seminar: VoiceBuilding for Text-to-Speech Synthesis
• Software project: Natural Language Processing with Neural Networks
• High-level Computer Vision (course work)
• Image Processing and Computer Vision (course work)
• Probabilistic Methods of Image Analysis (course work)
• Database Systems (course work)

Thesis:

• Thesis on developing a Neural Machine Translation model for a Neural TTS system for the integration of a Grapheme-to-Phoneme systhesis module with a Text(Phoneme)-to-Speech systhesis module.