ParsNets: A Parsimonious Composition of Orthogonal and Low-Rank Linear Networks for Zero-Shot Learning

Jingcai Guo*, Qihua Zhou, Xiaocheng Lu, Ruibin Li, Ziming Liu, Jie Zhang, Bo Han, Junyang Chen, Xin Xie, Song Guo

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

Abstract

This paper provides a novel parsimonious yet efficient design for zero-shot learning (ZSL), dubbed ParsNets, in which we are interested in learning a composition of on-device friendly linear networks, each with orthogonality and low-rankness properties, to achieve equivalent or better performance against deep models. Concretely, we first refactor the core module of ZSL, i.e., the visual-semantics mapping function, into several base linear networks that correspond to diverse components of the semantic space, wherein the complex nonlinearity can be collapsed into simple local linearities. Then, to facilitate the generalization of local linearities, we construct a maximal margin geometry on the learned features by enforcing low-rank constraints on intra-class samples and high-rank constraints on inter-class samples, resulting in orthogonal subspaces for different classes. To enhance the model's adaptability and counterbalance the over-/under-fittings, a set of sample-wise indicators is employed to select a sparse subset from these base linear networks to form a composite semantic predictor for each sample. Notably, maximal margin geometry can guarantee the diversity of features and, meanwhile, local linearities guarantee efficiency. Thus, our ParsNets can generalize better to unseen classes and can be deployed flexibly on resource-constrained devices.

Original languageEnglish
Title of host publicationProceedings of the 33rd International Joint Conference on Artificial Intelligence, IJCAI 2024
EditorsKate Larson
PublisherInternational Joint Conferences on Artificial Intelligence
Pages4062-4070
Number of pages9
ISBN (Electronic)9781956792041
DOIs
Publication statusPublished - 3 Aug 2024
Event33rd International Joint Conference on Artificial Intelligence, IJCAI 2024 - Jeju, Korea, Republic of
Duration: 3 Aug 20249 Aug 2024
https://ijcai24.org/ (Conference website)
https://ijcai24.org/whova-mobile-app/ (Conference program)
https://www.ijcai.org/Proceedings/2024/ (Conference proceedings)

Publication series

NameIJCAI International Joint Conference on Artificial Intelligence
ISSN (Print)1045-0823

Conference

Conference33rd International Joint Conference on Artificial Intelligence, IJCAI 2024
Country/TerritoryKorea, Republic of
CityJeju
Period3/08/249/08/24
Internet address

Scopus Subject Areas

  • Artificial Intelligence

User-Defined Keywords

  • Machine Learning: ML: Cost-sensitive learning
  • Machine Learning: ML: Ensemble methods
  • Machine Learning: ML: Few-shot learning
  • Machine Learning: ML: Learning sparse models

Cite this