View Selection
View selection, the process of choosing the most informative subset from a larger set of data points or options, is crucial for improving efficiency and performance in various applications. Current research focuses on developing sophisticated algorithms and models, such as those based on determinantal point processes, Bayesian methods, and contrastive learning, to optimize selection criteria across diverse domains including genetic algorithms, prompt engineering for LLMs, and neural rendering. These advancements are significant because effective view selection reduces computational costs, enhances model accuracy, and enables more robust and reliable decision-making in fields ranging from machine learning and computer vision to robotics and causal inference. The development of efficient and effective view selection methods is driving progress across numerous scientific disciplines and technological applications.
Papers
Selection, Ensemble, and Adaptation: Advancing Multi-Source-Free Domain Adaptation via Architecture Zoo
Jiangbo Pei, Ruizhe Li, Aidong Men, Yang Liu, Xiahai Zhuang, Qingchao Chen
Decoupling Weighing and Selecting for Integrating Multiple Graph Pre-training Tasks
Tianyu Fan, Lirong Wu, Yufei Huang, Haitao Lin, Cheng Tan, Zhangyang Gao, Stan Z. Li