Nithin Nagaraj’s research interests include Scientific Theories of Consciousness, Brain-inspired Artificial Intelligence, Causality, and Indic approaches to Mathematics, Computation, and Information Science.
Scientific Theories and Measures of Consciousness: Characterizing consciousness, the inner subjective feeling that is present in every experience, is a hard problem in neuroscience, but has important clinical implications. A leading neuro-scientific approach (Integrated Information Theory of Consciousness - IITC) to understanding consciousness is to measure the complex causal neural interactions in the brain. Elucidating the complex causal interplay between cortical neural interactions and the subsequent network computations is very challenging. We have proposed a novel scientific measure of consciousness - Network Causal Activity - using a Compression-Complexity Causality measure that can be applied to brain imaging measurements to quantify levels of consciousness.
Brain-inspired Artificial Intelligence: Chaos and Noise are ubiquitous in the Brain. Inspired by the chaotic firing of neurons and the constructive role of noise in neuronal models, we for the first time connect chaos, noise and learning. We have recently proposed Neurochaos Learning (NL) which uses chaotic neurons (unlike traditional artificial neural networks) and exploits noise to enhance learning (a phenomenon known as Stochastic Resonance). NL yields state-of-the-art performance in learning tasks and outperforms traditional Machine Learning/Deep Learning algorithms in the low training sample regime.
Causality: Determining and measuring cause-effect relationships is fundamental to most scientific studies of natural phenomena including the scientific study consciousness, especially brain-based measures of consciousness. The notion of causation is distinctly different from correlation which only looks at association of trends or patterns in measurements. Our group is interested in different notions of causality and focus especially on measuring causality from time series data. Causality testing finds numerous applications in diverse disciplines such as scientific theories and measures of consciousness, cognitive neuroscience, econometrics, climatology, physics, sustainability science, epidemiology, and artificial intelligence.
Journal Publications: Click here for the latest
Teaching: Please see the following link
Talks: Click here for the list
Patents: Click here
