Ashwin Gupta
Ashwin Gupta is an AI Systems Engineer at SkanAI, based in Bengaluru, India, where he works on scaling AI systems including LLM inference, RAG, and observability. He previously served as Head of Machine Learning at the Indian Institute of Science (IISc), leading ML and DL efforts for eVTOL design optimization. He is known for research in trajectory optimization and physics-informed neural networks, including work on stochastic multiple shooting trajectory optimization and model-based reinforcement learning with black-box dynamics.
“Stochastic Multiple Shooting Trajectory Optimization via Sequential Local Policy Evaluation”
Source→“On the cartpole swingup, only multiple shooting achieved convergence to the terminal set, with a mean terminal cost of 23.8 versus 130.3 for MPPI and 123.4 for single-shooting CEM (Table I).”
Source→“The authors 'show that we are able to synthesize approximate system Jacobians purely from rollouts, making the method suitable for model-based reinforcement learning with black-box dynamics' (Abstract).”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.