Multi-Domain Adversarial Learning
Skip to main content
eScholarship
Open Access Publications from the University of California

UCSF

UC San Francisco Previously Published Works bannerUCSF

Multi-Domain Adversarial Learning

  • Author(s): Schoenauer-Sebag, Alice
  • Heinrich, Louise
  • Schoenauer, Marc
  • Sebag, Michele
  • Wu, Lani F
  • Altschuler, Steve J
  • et al.
Abstract

Multi-domain learning (MDL) aims at obtaining a model with minimal average risk across multiple domains. Our empirical motivation is automated microscopy data, where cultured cells are imaged after being exposed to known and unknown chemical perturbations, and each dataset displays significant experimental bias. This paper presents a multi-domain adversarial learning approach, MuLANN, to leverage multiple datasets with overlapping but distinct class sets, in a semi-supervised setting. Our contributions include: i) a bound on the average- and worst-domain risk in MDL, obtained using the H-divergence; ii) a new loss to accommodate semi-supervised multi-domain learning and domain adaptation; iii) the experimental validation of the approach, improving on the state of the art on two standard image benchmarks, and a novel bioimage dataset, Cell.

Many UC-authored scholarly publications are freely available on this site because of the UC's open access policies. Let us know how this access is important for you.

Main Content
Current View