AI Bibliography
AI Bibliography |
|
 |
 |
 |
|
Wang, Y.-X., Ramanan, D., & Hebert, M. (2017). Learning to model the tail. Advances in Neural Information Processing Systems, 30. |
| Resource type: Journal Article BibTeX citation key: Wang2017 View all bibliographic details  |
Categories: Artificial Intelligence, Cognitive Science, Complexity Science, Computer Science, Data Sciences, Decision Theory, General, Mathematics Subcategories: AI transfer learning, Big data, Decision making, Deep learning, Forecasting, Machine learning, Machine recognition, Neural nets, Neurosymbolic, Q-learning Creators: Hebert, Ramanan, Wang Publisher: Collection: Advances in Neural Information Processing Systems |
| Attachments |
| Abstract |
|
We describe an approach to learning from long-tailed, imbalanced datasets that are prevalent in real-world settings. Here, the challenge is to learn accurate "few-shot'' models for classes in the tail of the class distribution, for which little data is available. We cast this problem as transfer learning, where knowledge from the data-rich classes in the head of the distribution is transferred to the data-poor classes in the tail. Our key insights are as follows. First, we propose to transfer meta-knowledge about learning-to-learn from the head classes. This knowledge is encoded with a meta-network that operates on the space of model parameters, that is trained to predict many-shot model parameters from few-shot model parameters. Second, we transfer this meta-knowledge in a progressive manner, from classes in the head to the "body'', and from the "body'' to the tail. That is, we transfer knowledge in a gradual fashion, regularizing meta-networks for few-shot regression with those trained with more training data. This allows our final network to capture a notion of model dynamics, that predicts how model parameters are likely to change as more training data is gradually added. We demonstrate results on image classification datasets (SUN, Places, and ImageNet) tuned for the long-tailed setting, that significantly outperform common heuristics, such as data resampling or reweighting.
|