Cross-Domain Few-Shot Learning by Representation Fusion

Thomas Adler, Johannes Brandstetter, Michael Widrich, Andreas Mayr, David Kreil, Michael K Kopp, Günter Klambauer, Sepp Hochreiter

Keywords: cross-domain learning, few-shot learning, Hebbian learning, ensemble learning, domain shift, domain adaptation, representation fusion

Abstract: In order to quickly adapt to new data, few-shot learning aims at learning from few examples, often by using already acquired knowledge. The new data often differs from the previously seen data due to a domain shift, that is, a change of the input-target distribution. While several methods perform well on small domain shifts like new target classes with similar inputs, larger domain shifts are still challenging. Large domain shifts may result in abstract concepts that are not shared between the original and the new domain. However, low-level concepts like edges in images might still be shared and useful. For cross-domain few-shot learning, we suggest representation fusion to unify different abstraction levels of a deep neural network into one representation. We propose Cross-domain Hebbian Ensemble Few-shot learning (CHEF), which consists of representation fusion by an ensemble of Hebbian learners acting on different layers of a deep neural network that was trained on the original domain. On the few-shot datasets miniImagenet and tieredImagenet, where the domain shift is small, CHEF is competitive with state-of-the-art methods. On cross-domain few-shot benchmark challenges with larger domain shifts, CHEF obtains state-of-the-art results in all categories. We further apply CHEF on a real-world cross-domain application in drug discovery. We consider a domain shift from bioactive molecules to environmental chemicals and drugs with twelve associated toxicity prediction tasks. On these tasks that are highly relevant for computational drug discovery, CHEF significantly outperforms all its competitors.

One-sentence Summary: We introduce the concept of representation fusion and, based on it, propose CHEF as a new method, which achieves new state-of-the-art results in cross-domain few-shot learning.

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

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:2010.06498/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=Dl-_XunJp7