Research map / Label-efficient and robust learning / Recognition and long-tail
Shared Space Transfer Learning for analyzing multi-site fMRI data
Tony Muhammad Yousefnezhad, Alessandro Selvitella, Daoqiang Zhang, Andrew Greenshaw, Russell Greiner
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SSTL global shared space on top of site-specific common features, versus raw MNI responses, MIDA and SIDeR | SSTL · Multi-site fMRI classification, overlapping datasets (A,B,C,G,H) | accuracy: p<0.05 in all 42 comparisons | Improved |
| compare SSTL with multi-view SRM, MDDL and MDMS | SSTL · Multi-site fMRI classification, overlapping datasets | accuracy | Improved |
| use single-view methods (MIDA, SIDeR) instead of raw neural responses | raw MNI-space classifier · Multi-site analysis with more than two sites | accuracy | Mixed |
| use multi-view techniques (SRM, MDDL, MDMS, SSTL) instead of single-view MIDA and SIDeR | transfer learning model · Multi-site fMRI classification, overlapping datasets | accuracy | Improved |
| apply SSTL versus MNI, MIDA and SIDeR on datasets with no shared subjects | SSTL · Multi-site fMRI classification, non-overlapping datasets | accuracy: p<0.05 in all 33 comparisons | Improved |
| use iterative multi-view optimization (MSMD, MDDL, SSA) instead of SSTL single-pass optimization | transfer learning methods · Runtime on overlapping dataset pairs | runtime | Hurt |
| use single-view methods instead of multi-view for runtime | transfer learning methods · Runtime on overlapping dataset pairs | runtime | Improved |
Closest papers by meaning
- Meta-Learning an In-Context Transformer Model of Human Higher Visual Cortex (NeurIPS 2025 · similarity 0.85)
- MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data (ICML 2024 · similarity 0.85)
- Enhancing Cross-Subject fMRI-to-Video Decoding with Global-Local Functional Alignment (ECCV 2024 · similarity 0.85)
- SwiFT: Swin 4D fMRI Transformer (NeurIPS 2023 · similarity 0.85)
- Mind Reader: Reconstructing complex images from brain activities (NeurIPS 2022 · similarity 0.84)
- MindAligner: Explicit Brain Functional Alignment for Cross-Subject Visual Decoding from Limited fMRI Data (ICML 2025 · similarity 0.83)
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.
