Understanding new tasks through the lens of training data via exponential tilting
Understanding new tasks through the lens of training data via exponential tilting
复制标题
DOI：
10.48550/arxiv.2205.13577
复制
发表时间：
2022-05
期刊：
影响因子：
--
通讯作者：
Subha Maity;M. Yurochkin;M. Banerjee;Yuekai Sun
中科院分区：
文献类型：
--
作者：
Subha Maity;M. Yurochkin;M. Banerjee;Yuekai Sun
尽管现代训练数据集规模庞大，但将机器学习模型部署到新任务中仍然是一个重大挑战。然而，可以想象的是，训练数据可以被重新加权，以更能代表新的（目标）任务。我们考虑重加权训练样本的问题，以深入了解目标任务的分布。具体来说，我们建立了一个基于指数倾斜假设的分布移位模型，并学习训练数据的重要性权重，使标记的训练和未标记的目标数据集之间的KL差异最小化。然后，学习到的训练数据权重可以用于下游任务，如目标性能评估、微调和模型选择。我们在水鸟和品种基准上证明了我们的方法的有效性。
Deploying machine learning models to new tasks is a major challenge despite the large size of the modern training datasets. However, it is conceivable that the training data can be reweighted to be more representative of the new (target) task. We consider the problem of reweighing the training samples to gain insights into the distribution of the target task. Specifically, we formulate a distribution shift model based on the exponential tilt assumption and learn train data importance weights minimizing the KL divergence between labeled train and unlabeled target datasets. The learned train data weights can then be used for downstream tasks such as target performance evaluation, fine-tuning, and model selection. We demonstrate the efficacy of our method on Waterbirds and Breeds benchmarks.