Research map / Label-efficient and robust learning / Domain adaptation and generalization
Principled Model Routing for Unknown Mixtures of Source Domains
Christoph Dann, Yishay Mansour, Teodor Vanislavov Marinov, Mehryar Mohri
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes principled routing of queries among domain-specialized models when the mixture of source domains is unknown and data for estimating it is limited or absent.
Problem
Selecting the best domain-specialized model for an unknown mixture of tasks, with little or no data to estimate that mixture, given that individual models' performance varies widely across domains.
Approach
Formulates model selection under an unknown mixture of source domains as a principled routing problem; the abstract is truncated so the specific method is not stated.
Closest papers by meaning
- Mixture Weight Estimation and Model Prediction in Multi-source Multi-target Domain Adaptation (NeurIPS 2023 · similarity 0.90)
- A Discriminative Technique for Multiple-Source Adaptation (ICML 2021 · similarity 0.89)
- STEM: An Approach to Multi-Source Domain Adaptation With Guarantees (ICCV 2021 · similarity 0.88)
- On Balancing Bias and Variance in Unsupervised Multi-Source-Free Domain Adaptation (ICML 2023 · similarity 0.88)
- Revisiting the Domain Shift and Sample Uncertainty in Multi-source Active Domain Transfer (CVPR 2024 · similarity 0.87)
- Domain2Vec: Vectorizing Datasets to Find the Optimal Data Mixture without Training (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
