Entity · technique

Paired Sampling for Domain Adaptation

techniqueactiveprovisionalpaired-sampling-for-domain-adaptation-e146988c·1 events·first seen 2d ago

Aliases: Paired Sampling for Domain Adaptation

More like this (12)

Any-Dimensional Learning by Sampling Automated Compliance Mapping in Cloud Security with Domain-Adapted Sentence Transformers Cross-sample Consistency Regularization finite-sample posterior sampling framework LoRA (Low-Rank Adaptation)Graph Sparse Sampling Force-Informed Re-Sampling Training LESS: Mutual-Stability Sampling for Diffusion Language Models Difference-Aware Retrieval Policies for Imitation Learning In-Context Reward Adaptation Sparse Subspace-to-Expert Sharing for Task-Agnostic Continual Learning Leveraging Instruction Tuning and Merging for Reasoning Model Adaptation

Recent events (1)

4arXiv · cs.LG·2d ago·source ↗

PSDA: Variance-reduced domain adaptation via paired sampling for distribution-matching objectives

A new arXiv preprint introduces Paired Sampling for Domain Adaptation (PSDA), a stochastic variance reduction technique designed for distribution-matching losses like correlation alignment and maximum mean discrepancy that lack finite-sum structure. PSDA constructs quadruplets of observations within and across domains by solving linear assignment problems to minimize expected gradient variance. Experiments on three domain shift datasets show improved target domain accuracy over related methods.

Evaluation and Benchmarking Paired Sampling for Domain Adaptation