针对表演性调整预训练骨干网络
Adjusting Pretrained Backbones for Performativity
AI总结：
针对深度学习模型部署引发的分布偏移与性能下降问题，提出模块化调整预训练骨干网络的方法，通过浅层适配器实现贝叶斯最优标签偏移校正，在视觉、语言任务的对抗采样场景中验证了效果，为深度学习表演性问题提供首个基准。
摘要：
随着深度学习模型的广泛部署，它们会以多种方式影响其所处环境，由此引发的分布偏移可能导致已部署模型出现意料之外的性能下降。现有预测表演性的方法通常会在预测未来结果时，将已部署模型的相关信息整合到特征向量中。这类方法虽具备颇具吸引力的理论特性，但修改预测任务的输入维度往往不具备实用性。\n为解决这一问题，我们提出一种新技术，以模块化方式针对表演性调整预训练骨干网络，既提升了样本效率，又能复用现有深度学习资产。我们聚焦表演性标签偏移场景，核心思路是训练一个浅层适配器模块，在给定待部署模型的充分统计量时，对骨干网络的logits（对数概率）执行贝叶斯最优标签偏移校正。由此，我们的框架将输入特定特征嵌入的构建过程，与支配表演性的机制解耦开来。\n受动态基准测试这一用例启发，我们在对抗采样设置下，针对视觉和语言任务评估了所提方法。结果表明，该方法能在再训练轨迹上实现更小的损失，还能帮助我们在候选模型中有效筛选，以预判性能下降问题。更广泛而言，我们的工作为解决深度学习中的表演性问题提供了首个基准。
英文摘要：
With the widespread deployment of deep learning models, they influence their environment in various ways. The induced distribution shifts can lead to unexpected performance degradation in deployed models. Existing methods to anticipate performativity typically incorporate information about the deployed model into the feature vector when predicting future outcomes. While enjoying appealing theoretical properties, modifying the input dimension of the prediction task is often not practical. To address this, we propose a novel technique to adjust pretrained backbones for performativity in a modular way, achieving better sample efficiency and enabling the reuse of existing deep learning assets. Focusing on performative label shift, the key idea is to train a shallow adapter module to perform a Bayes-optimal label shift correction to the backbone's logits given a sufficient statistic of the model to be deployed. As such, our framework decouples the construction of input-specific feature embeddings from the mechanism governing performativity. Motivated by dynamic benchmarking as a use-case, we evaluate our approach under adversarial sampling, for vision and language tasks. We show how it leads to smaller loss along the retraining trajectory and enables us to effectively select among candidate models to anticipate performance degradations. More broadly, our work provides a first baseline for addressing performativity in deep learning.
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。
