About

Rohan Kumar is an experienced machine learning researcher with a strong interest in algorithms and software development. He holds a Bachelor of Technology degree in Electrical and Electronics Engineering from the prestigious Indian Institute of Technology, Patna. With over 2 years of relevant experience, Rohan has worked as a Research Assistant at the AI-NLP-ML Research Lab at IIT Patna, where he worked on leveraging stance vs sentiment of a tweet, with respect to a given topic, to improve the quality of latent representation formed. He also worked on using attention mechanisms to improve the Political Stance detection accuracy and published a paper on the same at ICONIP 2019. Additionally, he worked on Multi-modal dialogue generation systems for multiparty dialogue generation. Rohan has also worked as a Research Intern at Greyc, France, where he was part of the HULTECH team, working on augmenting transformer networks with graph theory to generate multi-party conversations. He used Adversarial training in the form of shared-private networks for transfer learning, to create a robust dialogue generation system. He also created a new dataset to be used for such experiments with multi-party conversations, and the paper on the above work was accepted at IJCNN 2021. Currently, Rohan is working as a Software Engineer at toppr, where he is applying his skills in machine learning, AI, and software development to develop innovative solutions for the company. Rohan's technical skills include Software Engineering, ML, AI, Research Scientist, NLP, and test. He is a dedicated professional with a passion for delivering high-quality work.