Suman Ghosh
Research Interests
Humanity's ambition of building Artificial General Intelligence intrigues me immensely. Although I am an engineer by training and primarily work on applied AI, I am deeply interested in the theory of deep learning to build accountable and ethically sound AI. I have cultivated a recent interest in Computational Neuroscience to understand the theoretical basis of cognition, memory and learning.
I have been involved in theoretical and applied interdisciplinary research in:
-
Learning (deep learning, spiking neural networks, reinforcement learning, self-supervised learning)
-
Vision (visual attention, structure from motion, visual odometry, SLAM, stereo disparity, OCR)
-
Robotics (human robot interaction, neuromorphic perception, motion and path planning)
-
Human Computer Interaction (assistive interfaces)
-
Image Processing (contrast enhancement, scratched image restoration)
-
Evolutionary Computing (differential evolution, particle swarm optimization)