New Formulation
Recent research on novel formulations focuses on improving the efficiency and accuracy of various computational tasks by redefining problem structures and incorporating advanced techniques. Key areas of investigation include developing instance-mask-based approaches for improved geometric representation, leveraging quantum computing for optimization problems, and employing causal inference and robust training strategies to handle noisy data and uncertainty. These advancements are significant for enhancing the performance of machine learning models in diverse applications, ranging from road topology mapping and recommendation systems to malware detection and materials science.
Papers
Thales: Formulating and Estimating Architectural Vulnerability Factors for DNN Accelerators
Abhishek Tyagi, Yiming Gan, Shaoshan Liu, Bo Yu, Paul Whatmough, Yuhao Zhu
Formulation of problems of combinatorial optimization for solving problems of management and planning of cloud production
M. V. Saramud, E. A. Spirin, E. P. Talay, I. I. Pikalov