Sravan Bodapati

Papers

Total Citations

H-Index

About

Sravan Bodapati is a researcher advancing the frontiers of reinforcement learning and robotics, with a focus on bridging the critical gap between simulation and real-world deployment. His most-cited work, "Zero-Shot Reinforcement Learning with Deep Attention Convolutional Neural Networks" (2020, 7 citations), tackles the persistent challenge of simulation-to-real world transfer by leveraging attention mechanisms within deep convolutional architectures. This innovative approach enables neural network models to generalize across simulation-to-simulation and simulation-to-real world environments without requiring additional training data, effectively closing the "reality gap" that has long hindered robotic applications. Rather than relying on traditional domain adaptation or decoupling perception and dynamics, Bodapati's method demonstrates how attention-based models can learn robust representations that transfer seamlessly to novel physical settings. His contributions are particularly significant for autonomous systems where collecting real-world training data is expensive or dangerous. With growing recognition in the reinforcement learning community, Bodapati's work continues to influence how researchers approach zero-shot transfer learning, offering a more efficient pathway toward deploying intelligent agents in complex, unstructured environments.