Robust Performance
Robust performance in various applications, from AI model training to robotic control, focuses on developing systems that maintain high accuracy and efficiency despite uncertainties or disturbances. Current research emphasizes methods like adaptive loss functions, Bayesian optimization, and novel model architectures (e.g., incorporating Computing-in-Memory architectures) to enhance robustness against noise, outliers, and variations in data size or quality. These advancements are crucial for deploying reliable AI systems in real-world settings and improving the efficiency and accuracy of complex simulations and control systems across diverse scientific and engineering domains.
19papers
Papers
December 1, 2024
September 26, 2024
January 21, 2024
October 14, 2023
December 19, 2022
November 24, 2022
December 7, 2021