Selection Policy
Selection policy, encompassing the strategic choice of data, individuals, or actions based on specific criteria, aims to optimize outcomes in diverse fields from machine learning to evolutionary algorithms and even social decision-making. Current research emphasizes efficient algorithms for selecting subsets of data (e.g., using uncertainty sampling or diversity-based methods) to improve model training and performance, particularly in resource-constrained environments. These advancements are crucial for enhancing the efficiency and fairness of various applications, ranging from large language model deployment to mitigating bias in algorithmic decision-making. The development of robust and adaptable selection policies is vital for addressing challenges in scalability, computational cost, and ethical considerations across numerous domains.
Papers
Efficient Subgraph GNNs by Learning Effective Selection Policies
Beatrice Bevilacqua, Moshe Eliasof, Eli Meirom, Bruno Ribeiro, Haggai Maron
Which Examples to Annotate for In-Context Learning? Towards Effective and Efficient Selection
Costas Mavromatis, Balasubramaniam Srinivasan, Zhengyuan Shen, Jiani Zhang, Huzefa Rangwala, Christos Faloutsos, George Karypis