Two new research papers propose methods to reduce variance in domain adaptation techniques. The first paper, "Variance-reduced Domain Adaptation using Paired Sampling" (PSDA), introduces a stochastic variance reduction (SVR) technique that pairs observations within and across domains to minimize gradient variance. The second paper, "Online Variance Reduction for Domain Adaptation on Streaming Data," presents ARROW, an adaptive SVR algorithm designed for streamed data that maintains moving average references and adaptively reweights minibatches. AI
IMPACT These methods aim to improve the accuracy and efficiency of machine learning models in handling data from different distributions.
RANK_REASON Two academic papers published on arXiv proposing novel methods for domain adaptation.
- Arrow
- arXiv
- correlation alignment
- Maximum Mean Discrepancy
- Paired Sampling for Domain Adaptation
- Stochastic variance reduction
- Unsupervised Domain Adaptation in Brain Lesion Segmentation with Adversarial Networks
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
