Sakyasingha Dasgupta
Bernstein Center for Computational Neuroscience Göttingen, University of Göttingen
Papers
13
Total Citations
297
H-Index
9
About
Sakyasingha Dasgupta is a computational neuroscientist and robotics researcher whose work bridges biological intelligence and machine learning, with particular expertise in neural computation, adaptive locomotion control, and deep learning for robotics. His research draws inspiration from neuroscience to develop bio-inspired algorithms capable of enabling robust, adaptive behavior in artificial systems. Among his most influential contributions is his work on chaotic central pattern generators (CPGs) for legged locomotion, demonstrating how chaos-based neural controllers can generate flexible walking patterns and compensate for limb damage in hexapod robots — work that has accumulated over 50 citations. Dasgupta has also made significant strides in reservoir computing and information dynamics, developing self-adaptive systems for temporal memory tasks, and in understanding how Hebbian cell assemblies enable nonlinear computation in neural networks. More recently, he has advanced the field of sim-to-real transfer learning, leveraging variational autoencoders to bridge the gap between synthetic and real-world imagery for robotic applications, garnering nearly 40 citations. His research on neuromodulatory learning systems integrating cerebellar and basal ganglia mechanisms further highlights his commitment to biologically grounded approaches to intelligent behavior.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3The Use of Hebbian Cell Assemblies for Nonlinear Computation42 citations · 2015
- 4
- 5
- 6
- 7
- 8
- 9
- 10