Rachit Rajat

Papers

Total Citations

H-Index

About

Rachit Rajat is a researcher at the forefront of hardware-accelerated artificial intelligence, with a primary focus on efficient implementations of reinforcement learning algorithms. His most impactful work centers on QTAccel, a pioneering FPGA-based design for Q-Table Reinforcement Learning (QRL) accelerators, which has garnered 27 combined citations. Rajat’s key contribution lies in demonstrating that QRL, which iteratively improves state-action value estimates stored in a table, can be dramatically accelerated on reconfigurable hardware. This approach offers a compelling alternative to neural network-based methods for tractable state spaces, achieving superior performance and energy efficiency. By bridging the gap between classical reinforcement learning and modern hardware design, Rajat’s work provides a practical pathway for deploying AI in resource-constrained environments, such as embedded systems and real-time control applications. His research highlights the untapped potential of table-based learning methods when paired with custom digital logic, making him a notable figure in the intersection of AI algorithms and hardware acceleration.