Multidisciplinary teams are essential for machine learning research says doctoral researcher Parthasaarathy Sudarsanam

“To be honest, I did not really have a clear direction of what I wanted to do even after completing my bachelor's degree,” admits Parthasaarathy Sudarsanam, a 28-year-old doctoral researcher from Chennai, India. It was his tenure at Sony, working on machine learning and signal processing projects, that ignited his passion for the field. This pivotal moment set him on a path that led to Tampere University, where he now engages in the groundbreaking ‘CONVERGENCE of Humans and Machines’ project.

Sudarsanam has studied Electronics and Communication Engineering at Anna University and Machine Learning and Signal Processing at Tampere University. His current research focuses on socially aware machine perception, a natural extension of his earlier work in multimodal machine perception, which combines audio and language modalities. 

Photo: Antti Yrjönen

“Technology has already created a profound impact on various aspects of human life, including communication and education. I am intrigued by the potential benefits and challenges that emerge when technology and humanity converge.”

The rapid learning curve at Sony and the capabilities of the machine learning models he worked on convinced him of his path. “The feeling that the code I wrote could generate such capabilities made me decide that this is what I am going to work on,” he recalls.

When it comes to explaining his research to those unfamiliar with the field, Sudarsanam likens it to personalising media experiences. “Each of us interprets movies, songs, or novels differently based on our culture, language, and experiences. My research aims to modify the behaviour of machine learning models to suit individual characteristics, creating a highly personalised AI for each user,” he explains.

The real-world applications of Sudarsanam’s research are vast. Imagine education apps or fitness apps tailored to individual needs and habits, or a ChatGPT that understands and adapts to your unique way of thinking. 

We are laying the foundation for personalised AI

Parthasaarathy Sudarsanam

In the future, technology will be better adapted to our individual needs and habits. “We will hopefully reach an equilibrium of human-machine interaction, where technology blends smoothly into our lives,” Sudarsanam predicts. “I think the future possibilities are endless.”

Photo: Antti Yrjönen

The multidisciplinary environment of the CONVERGENCE project is essential, according to Sudarsanam. Working together with researchers from both technology and the humanities ensures that the technology developed works in harmony with people. 

“Not just for this Convergence project, I think it is time all machine learning-related research starts having multidisciplinary teams. With the capabilities of LLMs and other generative models, we are already starting to see how critical people are towards the possible hazards of these technologies.”

Photo: Antti Yrjönen

Each CONVERGENCE doctoral researcher has two main supervisors. The main supervisors of Sudarsanam are Tuomas Virtanen and Maija Hirvonen, and the other supervisors are Esa Rahtu and Okko Räsänen.