Co Design
Co-design is a synergistic approach that simultaneously optimizes both the algorithm and hardware components of a system, aiming to achieve superior performance and efficiency compared to independent optimization. Current research focuses on diverse applications, including AI accelerators, robotic control, and large language model deployment, employing techniques like reinforcement learning, Bayesian optimization, and neural architecture search to achieve this co-optimization. This interdisciplinary field is significantly impacting various domains by enabling the development of more efficient, robust, and cost-effective systems across diverse applications, from edge computing to medical AI.
Papers
Data-Algorithm-Architecture Co-Optimization for Fair Neural Networks on Skin Lesion Dataset
Yi Sheng, Junhuan Yang, Jinyang Li, James Alaina, Xiaowei Xu, Yiyu Shi, Jingtong Hu, Weiwen Jiang, Lei Yang
Automated and Holistic Co-design of Neural Networks and ASICs for Enabling In-Pixel Intelligence
Shubha R. Kharel, Prashansa Mukim, Piotr Maj, Grzegorz W. Deptuch, Shinjae Yoo, Yihui Ren, Soumyajit Mandal