Interview with TERAIS’ new postdoctoral researcher Laxmi R Iyer, PhD

Welcome to an interview with our newest colleague at the TERAIS project, Laxmi R. Iyer, PhD, post doctoral researcher with a lot of experience in computational cognitive modelling and neuromorphic engineering. Laxmi brings her expertise to the TERAIS team at Comenius University in Bratislava, Slovakia, where she will be focusing on advancing artificial intelligence and robotics through biologically inspired computational models.

Laxmi’s role at TERAIS will not only enhance the research capacity of DAI UKBA but also strengthen the project’s commitment to research excellence, international collaboration, and innovative breakthroughs in AI and robotics. Join us as we explore her career journey, research ambitions, and her contributions to shaping the future of technology and cognitive science.

Can you tell us about your academic and research background and how it led you to become involved in the TERAIS project as a postdoctoral researcher?

In computational cognitive modelling, I have worked with models of creativity and ideation, spatial memory, perception and prediction of actions, and semantic cognition. In neuromorphic engineering, I have worked on leaky integrate-and-fire (LIF) spiking neurons and the spike-timing dependent plasticity (STDP) learning rule. I have analysed the characteristics of neuromorphic datasets such as N-MNIST and DVS-Gestures.

I was interested in joining the TERAIS project due to several factors. First was the research profiles of the Cognitive Science researchers at UKBA. Second was the international collaboration opportunities that TERAIS project offers with the Italian Institute of Technology and University of Hamburg. Thirdly, the Cognitive Science Master’s program that UKBA offers in collaboration with other universities was also impressive. It suggested to me that this was a fertile ecosystem for Cognitive Science research, and a very fruitful opportunity for me to advance in my career. I was also excited about Cognitive Robotics - the idea of using humanoid robots as a testbed for cognitive theories.

Can you share what initially motivated you to pursue a Ph.D. in Computer Science, particularly with a focus on connectionist models of creativity and ideation?

I did my B.Eng. in Computer Engineering, but was drawn more to Computer Science courses, particularly programming, mathematical logic and AI. I enrolled in a Ph.D. program in the field of AI. At the beginning of my Ph.D. I spoke to my supervisor about the research he was doing. He gave me several papers to read to give a flavour of his research interests. I was most interested in using neural networks to model cognitive phenomena. He informed me that he had just started some preliminary work on a project on idea generation. This is how I got into this project. It was a combination of interest and opportunity.

With advancements in AI and machine learning, where do you think computational models of the brain will have the most significant impact over the next decade?

If you take a look at the history of AI, initially AI was all about logical IF-ELSE rules, procedural programming and expert systems. Today AI is mostly done by neural networks and learning, which is much more biological. Researchers are looking towards neuromorphic systems - spiking neural network models and neuromorphic hardware to develop efficient AI solutions. In so many different ways, at each step, AI looks towards cognitive science. Currently AI relies on intelligent mathematical solutions to various issues in intelligence. Cognitive science takes a different approach, which is to understand how the brain functions. As our understanding of the brain matures, I believe that cognitive science will continue to be incorporated into AI. The reason for this is that our brain is the best AI system there is. In some cases, AI performance has exceeded human performance. But mostly, the human brain is the gold standard for AI research and applications.

What specific areas of artificial intelligence are you most passionate about, and how do they align with the goals of the TERAIS project?

I am most passionate about the computational modelling of language in the brain. I feel that the area that connects most clearly to robotics is sensorimotor grounding of language. NLP has had a long history and most language models rely on association between different words to one another. In sharp contrast, cognitive scientists argue that we acquire language through the sensory and motor interaction of our body with the environment. As robots are embodied, they provide a unique opportunity to study sensorimotor grounding in language.

How do you think the TERAIS project will impact your career trajectory and future opportunities in the field of robotics and AI?

As I had mentioned earlier, I feel this is a very useful opportunity for advancing in my career in several ways - firstly, the interaction with excellent researchers in my field will broaden my own thinking. Secondly, the collaboration opportunities with excellent researchers in other countries may prove very useful now and in the future. Finally, the Cognitive Science Masters program at UKBA in collaboration with other universities can provide rich interactions, discussions, and possibly excellent students that I can potentially work with.

What advice would you give to students and early-career researchers who are interested in pursuing AI and cognitive science?

To students pursuing AI research, I would firstly tell them to make sure research is what they want. I would ask them to identify what topics they want and who they want to work with.

To early-career researchers I would say what I have been told - choose an area of interest and become very strong in that area. Read papers, as much as you can - as there is no shortcut to reading. Plan your research, systematically. As a researcher, you are like an entrepreneur. You chart your own path. Open your mind to collaborations and interactions. As an AI researcher, familiarise yourself with Machine Learning frameworks and libraries, and keep in touch with the latest technology in implementing AI.

To researchers pursuing cognitive science, I would say be open to symposia and books on computational modelling of cognitive phenomena. Look for opportunities to work with excellent researchers in your area of interests.