English

Contextual HyperNetworks for Novel Feature Adaptation

Machine Learning 2021-04-14 v1

Abstract

While deep learning has obtained state-of-the-art results in many applications, the adaptation of neural network architectures to incorporate new output features remains a challenge, as neural networks are commonly trained to produce a fixed output dimension. This issue is particularly severe in online learning settings, where new output features, such as items in a recommender system, are added continually with few or no associated observations. As such, methods for adapting neural networks to novel features which are both time and data-efficient are desired. To address this, we propose the Contextual HyperNetwork (CHN), an auxiliary model which generates parameters for extending the base model to a new feature, by utilizing both existing data as well as any observations and/or metadata associated with the new feature. At prediction time, the CHN requires only a single forward pass through a neural network, yielding a significant speed-up when compared to re-training and fine-tuning approaches. To assess the performance of CHNs, we use a CHN to augment a partial variational autoencoder (P-VAE), a deep generative model which can impute the values of missing features in sparsely-observed data. We show that this system obtains improved few-shot learning performance for novel features over existing imputation and meta-learning baselines across recommender systems, e-learning, and healthcare tasks.

Keywords

Cite

@article{arxiv.2104.05860,
  title  = {Contextual HyperNetworks for Novel Feature Adaptation},
  author = {Angus Lamb and Evgeny Saveliev and Yingzhen Li and Sebastian Tschiatschek and Camilla Longden and Simon Woodhead and José Miguel Hernández-Lobato and Richard E. Turner and Pashmina Cameron and Cheng Zhang},
  journal= {arXiv preprint arXiv:2104.05860},
  year   = {2021}
}

Comments

17 pages, 9 Figures, workshop paper at NeurIPS 2020 Meta-Learning Workshop

R2 v1 2026-06-24T01:06:10.000Z