Domain Generalization
A research direction in machine learning requiring a model to maintain performance on out-of-distribution target domains whose data is entirely unseen during training, posing greater challenges than conventional unsupervised domain adaptation.
Timeline (last 90 days)
The leave-batch-out setting of BatchBench is closer to domain generalization, i.e., the model never sees the target batch during training.
Domain generalization requires models to generalize to completely unseen target domain data, with commonly used methods including meta-learning, causal inference-based invariant feature learning, and multi-source domain data augmentation.
All Facts (2)
The leave-batch-out setting of BatchBench is closer to domain generalization, i.e., the model never sees the target batch during training.
50%UnverifiedDomain generalization requires models to generalize to completely unseen target domain data, with commonly used methods including meta-learning, causal inference-based invariant feature learning, and multi-source domain data augmentation.
50%