Interpretable Domain Adaptation for Hidden Subdomain Alignment in the Context of Pre-trained Source Models - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2022

Interpretable Domain Adaptation for Hidden Subdomain Alignment in the Context of Pre-trained Source Models

Pascal Germain
Christophe Biernacki
  • Function : Author
  • PersonId : 923939

Abstract

Domain adaptation aims to leverage source domain knowledge to predict target domain labels. Most domain adaptation methods tackle a single-source, single-target scenario, whereas source and target domain data can often be subdivided into data from different distributions in real-life applications (e.g., when the distribution of the collected data changes with time). However, such subdomains are rarely given and should be discovered automatically. To this end, some recent domain adaptation works seek separations of hidden subdomains, w.r.t. a known or fixed number of subdomains. In contrast, this paper introduces a new subdomain combination method that leverages a variable number of subdomains. Precisely, we propose to use an inter-subdomain divergence maximization criterion to exploit hidden subdomains. Besides, our proposition stands in a target-to-source domain adaptation scenario, where one exploits a pre-trained source model as a black box; thus, the proposed method is model-agnostic. By providing interpretability at two complementary levels (transformation and subdomain levels), our method can also be easily interpreted by practitioners with or without machine learning backgrounds. Experimental results over two fraud detection datasets demonstrate the efficiency of our method.
Fichier principal
Vignette du fichier
2006.ZhangL.pdf (377.51 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03505639 , version 1 (31-12-2021)

Identifiers

  • HAL Id : hal-03505639 , version 1

Cite

Luxin Zhang, Pascal Germain, Yacine Kessaci, Christophe Biernacki. Interpretable Domain Adaptation for Hidden Subdomain Alignment in the Context of Pre-trained Source Models. 36th AAAI Conférence on Artificial Intelligence, Feb 2022, Vancouver, Canada. ⟨hal-03505639⟩
195 View
201 Download

Share

Gmail Facebook X LinkedIn More