Exploiting a Zoo of Checkpoints for Unseen Tasks

Abstract

There are so many models in the literature that it is difficult for practitioners to decide which combinations are likely to be effective for a new task. This paper attempts to address this question by capturing relationships among checkpoints published on the web. We model the space of tasks as a Gaussian process. The covariance can be estimated from checkpoints and unlabeled probing data. With the Gaussian process, we can identify representative checkpoints by a maximum mutual information criterion. This objective is submodular. A greedy method identifies representatives that are likely to "cover" the task space. These representatives generalize to new tasks with superior performance. Empirical evidence is provided for applications from both computational linguistics as well as computer vision.

arXiv commentAccepted in Neurips 2021

References in corpus (12)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Language Models are Few-Shot Learners
• Popular Ensemble Methods: An Empirical Study
• Longformer: The Long-Document Transformer
• Similarity of Neural Network Representations Revisited
• Algorithms for Learning Kernels Based on Centered Alignment
• Meta-Learning by Adjusting Priors Based on Extended PAC-Bayes Theory
• LEEP: A New Measure to Evaluate Transferability of Learned Representations
• Semantic Tagging with Deep Residual Networks
• Deep Model Transferability from Attribution Maps
• Ranking Neural Checkpoints