Aishvarya shree V G

I’m an AI Researcher and Ph.D. researcher specializing in Machine Learning, Deep Learning, Explainable AI (XAI), anomaly detection, and trustworthy AI. I design, evaluate, and interpret AI systems to make predictions more transparent and reliable.

In my RUSA-funded research as a Research Assistant, I developed deep learning models such as LSTM Autoencoders and XGBoost for anomaly detection across cybersecurity and healthcare datasets. I build end-to-end ML pipelines (data preprocessing, feature engineering, validation, and performance evaluation), run comparative experiments with hyperparameter tuning and ablation studies, and assess robustness under noisy and imbalanced data.

I also implement SHAP-based explainable AI to provide feature-level interpretation, and I study computational efficiency and inference latency for deployment in resource-constrained environments. My work includes reproducible benchmarking pipelines, interactive dashboards (Tableau, Power BI), publications in Scopus/IEEE, and a patent application focused on enhanced security for healthcare data integration.