Deep Ensembles for Low-Data Transfer LearningDownload PDF

28 Sept 2020 (modified: 22 Oct 2023)ICLR 2021 Conference Blind SubmissionReaders: Everyone
Keywords: transfer learning, representation learning, computer vision, ensembles
Abstract: In the low-data regime, it is difficult to train good supervised models from scratch. Instead practitioners turn to pre-trained models, leveraging transfer learning. Ensembling is an empirically and theoretically appealing way to construct powerful predictive models, but the predominant approach of training multiple deep networks with different random initialisations collides with the need for transfer via pre-trained weights. In this work, we study different ways of creating ensembles from pre-trained models. We show that the nature of pre-training itself is a performant source of diversity, and propose a practical algorithm that efficiently identifies a subset of pre-trained models for any downstream dataset. The approach is simple: Use nearest-neighbour accuracy to rank pre-trained models, fine-tune the best ones with a small hyperparameter sweep, and greedily construct an ensemble to minimise validation cross-entropy. When evaluated together with strong baselines on 19 different downstream tasks (the Visual Task Adaptation Benchmark), this achieves state-of-the-art performance at a much lower inference budget, even when selecting from over 2,000 pre-trained models. We also assess our ensembles on ImageNet variants and show improved robustness to distribution shift.
One-sentence Summary: A study of ensembling pre-trained models for downstream computer vision classification tasks with minimal data, and a proposal for an approach that utilises diversity from pre-training to improve accuracy and robustness.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Supplementary Material: zip
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:2010.06866/code)
Reviewed Version (pdf): https://openreview.net/references/pdf?id=KSwfRwvzrd
9 Replies

Loading