DANCE: 双重自适应邻域符合估计
DANCE: Doubly Adaptive Neighborhood Conformal Estimation
AI总结：
DANCE通过结合双重局部自适应最近邻方法，提升预训练模型在不确定性量化中的集大小效率和鲁棒性。
摘要：
近期复杂深度学习模型的发展带来了在多种数据表示类型上准确预测的前所未有的能力。为了对这些模型进行不确定性量化，符合预测方法因其适应性和统计有效性而受到关注。在分类任务中，符合方法通常专注于利用logit分数。然而，对于预训练模型而言，如果不针对目标任务进行校准，这可能导致效率低下且过于保守的集大小。我们提出了DANCE，一种基于双重局部自适应最近邻的符合算法，直接使用数据嵌入表示来结合两个新颖的非符合分数。DANCE首先从嵌入层拟合一个任务自适应的核回归模型，然后利用学习的核空间生成最终的预测集以进行不确定性量化。我们测试了最先进的局部、任务适应和零样本符合基线，证明了DANCE在各种数据集上集大小效率和鲁棒性方面的优越性。
英文摘要：
The recent developments of complex deep learning models have led to unprecedented ability to accurately predict across multiple data representation types. Conformal prediction for uncertainty quantification of these models has risen in popularity, providing adaptive, statistically-valid prediction sets. For classification tasks, conformal methods have typically focused on utilizing logit scores. For pre-trained models, however, this can result in inefficient, overly conservative set sizes when not calibrated towards the target task. We propose DANCE, a doubly locally adaptive nearest-neighbor based conformal algorithm combining two novel nonconformity scores directly using the data's embedded representation. DANCE first fits a task-adaptive kernel regression model from the embedding layer before using the learned kernel space to produce the final prediction sets for uncertainty quantification. We test against state-of-the-art local, task-adapted and zero-shot conformal baselines, demonstrating DANCE's superior blend of set size efficiency and robustness across various datasets.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
