Right Problem
"Right Problem" research focuses on identifying and effectively addressing the core challenges within various problem domains, ensuring that solutions are tailored to the specific needs and constraints of the problem. Current research emphasizes developing and evaluating appropriate performance metrics, employing advanced algorithms like deep learning and reinforcement learning to solve complex problems in diverse fields such as image analysis, robotics, and natural language processing. This work is crucial for advancing scientific understanding and improving the efficacy of practical applications by ensuring that computational resources are directed towards the most impactful and relevant aspects of a given problem.
Papers
Tackling the Problem of Distributional Shifts: Correcting Misspecified, High-Dimensional Data-Driven Priors for Inverse Problems
Gabriel Missael Barco, Alexandre Adam, Connor Stone, Yashar Hezaveh, Laurence Perreault-Levasseur
Revising the Problem of Partial Labels from the Perspective of CNNs' Robustness
Xin Zhang, Yuqi Song, Wyatt McCurdy, Xiaofeng Wang, Fei Zuo