Lasitha Vidyaratne

Old Dominion University, Tennessee State University

Papers

Total Citations

H-Index

About

Lasitha Vidyaratne is a researcher specializing in deep learning, computer vision, and neural network architectures, with a particular focus on applying these technologies to humanoid robotics and recognition systems. His work has made notable contributions to the practical deployment of artificial intelligence in real-world robotic platforms, most prominently the NAO humanoid robot. Vidyaratne's most recognized contribution explores the use of Deep Spiking Recurrent Networks (SRN) for robust object recognition, demonstrating how biologically inspired neural architectures can overcome computational limitations while maintaining strong performance — a paper that has garnered 10 citations. His subsequent investigation into convolutional neural network transfer learning for face recognition further established his expertise in efficient model adaptation, showing how pre-trained architectures can be effectively repurposed without training from scratch, earning 8 citations. His 2020 survey on deep neural networks in speech and vision systems reflects his broader commitment to synthesizing advancements across multiple AI domains, making complex research accessible to the wider scientific community. Collectively, Vidyaratne's work bridges theoretical deep learning research with tangible robotics applications, offering valuable insights for researchers working at the intersection of machine perception, transfer learning, and autonomous systems.