Our April issue brings together diverse topics of interest:

• Computer Audition: From Task-Specific Machine Learning to Foundation Models
• Talkative Power Conversion
• Federated Domain Generalization

Check out the full issue: bit.ly/4nqNt6X

#domaingeneralization #machinelearning #powerconversion

#xAILabBamberg #BaCAI #HightechAgendaBayern #MedicalImageAnalysis #DomainGeneralization #Robustness #Reliability
@uni-bamberg.de @miccaisociety.bsky.social

MS‑UDG learns minimal sufficient semantic representations without any labels and significantly beats prior SSL and UDG methods on multiple benchmarks. Read more: https://getnews.me/minimal-semantic-sufficiency-boosts-unsupervised-domain-generalization/ #unsupervised #domaingeneralization

CI‑TTA improves domain‑generalization, with experiments on PACS and Office‑Home benchmarks showing consistent accuracy gains across DG algorithms and backbones. https://getnews.me/class-invariant-test-time-augmentation-boosts-domain-generalization/ #domaingeneralization #testtimeaugmentation

If you are presenting a poster, please already set it up between 2PM and 2.30PM.

See you in Turin!

#ecml #ecmlpkdd #machinelearning #domainadaptation #domaingeneralization #trustworthyai #continuallearning

Federated domain generalization combines federated learning and domain generalization to enable collaborative model training across diverse sources, ensuring effective generalization to unseen domains while preserving data privacy. Lear more at: bit.ly/4npiWH8

#domaingeneralization #MachineLearning