Faculty Fellowship for conducting AI Model Optimization Research for Inference Acceleration in Edge Computing Environments

The Army Research Laboratory Research Associateship Program (ARL-RAP) Faculty Fellowship opportunity focuses on developing theoretical and experimental approaches for generalized AI model inference acceleration on resource-constrained heterogeneous edge computing platforms. The research aims to predict optimal AI model architecture through neural network architecture search (NAS) to achieve expected inference acceleration, covering both convolutional neural networks and large language models. The fellowship is designed to significantly increase the involvement of creative and highly trained scientists and engineers from academia and industry in scientific and technical areas of interest and relevance to the Army. Participants work with ARL scientists and engineers who help shape and execute the Army's program for meeting the challenge of developing technologies that will support Army forces in meeting future operational needs. The research covers developing mathematical models to understand trade-offs between accuracy, latency, and compression of optimized AI models; investigating state-of-the-art quantization and model pruning approaches; formulating mathematical theoretical foundations to guide optimization; and developing layer-wise gradual optimization approaches.