Hardik Raina

Solely responsible for conducting research and curating datasets for the application of generative models in computer vision. Successfully handled challenges related to insufficient information and dataset rebalancing. Utilized Natural Language Processing (NLP) techniques to improve Recommender Systems performance. Devised personality matrices to enhance accuracy scores on sparse datasets, including effectively addressing cold‑start problems. Achieved an accuracy improvement of over 25.8% compared to traditional models. Demonstrated expertise in Artificial Intelligence (AI) by creating a Deep Convolutional Generative Adversarial Network (DCGAN) architecture. Conducted extensive research on progressive model enhancements through Online Learning and Multi‑Armed Bandit (MAB) approaches.