Used reinforcement learning to optimize the design of electro-photonic accelerators to support a. DNN workloads associated with autonomous vehicles such as pedestrian detection, traffic light recognition, and trajectory prediction.b. training and inference for Hyperdimensional Computing (HDC) algorithms for image classification, speech recognition, and time series analysis. HDC has recently emerged as a novel brain-inspired machine learning (ML) technique whose accuracy is close to DNNs in various applications, but it requires fewer computational resources than DNNs.