Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published November 2019 | Submitted
Book Section - Chapter Open

Task2Vec: Task Embedding for Meta-Learning

Abstract

We introduce a method to generate vectorial representations of visual classification tasks which can be used to reason about the nature of those tasks and their relations. Given a dataset with ground-truth labels and a loss function, we process images through a "probe network" and compute an embedding based on estimates of the Fisher information matrix associated with the probe network parameters. This provides a fixed-dimensional embedding of the task that is independent of details such as the number of classes and requires no understanding of the class label semantics. We demonstrate that this embedding is capable of predicting task similarities that match our intuition about semantic and taxonomic relations between different visual tasks. We demonstrate the practical value of this framework for the meta-task of selecting a pre-trained feature extractor for a novel task. We present a simple meta-learning framework for learning a metric on embeddings that is capable of predicting which feature extractors will perform well on which task. Selecting a feature extractor with task embedding yields performance close to the best available feature extractor, with substantially less computational effort than exhaustively training and evaluating all available models.

Additional Information

© 2019 IEEE.

Attached Files

Submitted - 1902.03545.pdf

Files

1902.03545.pdf
Files (1.6 MB)
Name Size Download all
md5:d045abca98f29f660f08a8b4729a5955
1.6 MB Preview Download

Additional details

Created:
August 19, 2023
Modified:
October 20, 2023