About Me
I am Yuan Ma from Beijing, China. I am a Ph.D. student at Institute Of Computing Technology, Chinese Academy of Sciences, advised by Prof. Yiqiang Chen, and I work primarily in the intersection of transfer learning algorithms and applications in ubiquitous computing.
Research Interests:
- Transfer Learning： Domain Adaptation (DA) and Generalization (DG)
- Retrieval-Augmented Generation (RAG)
- AI for Healthcare
Publications
You can also find my articles on my Google Scholar profile.
Publications
DDIR: Domain-Disentangled Invariant Representation Learning for Tailored Predictions
Knowledge-Based Systems 2025
A domain decoupling invariant representation learning method is proposed, which effectively separates DOI and domain-invariant features to achieve accurate predictions in cross-domain scenarios.
DDIR method
Grade-Skewed Domain Adaptation via Asymmetric Bi-Classifier Discrepancy Minimization for Diabetic Retinopathy Grading
IEEE Transactions on Medical Imaging 2024
To address the issue of grade skew in medical image analysis, an asymmetric dual classifier method was developed, which significantly improved the diagnostic accuracy.
ABiD grading system
Class-imbalanced time series domain adaptation via multi-expert consistency entropy minimization
IEEE International Conference on Bioinformatics and Biomedicine 2024
By adopting the multi-expert consistency entropy minimization strategy, we effectively address the dual challenges of the reliability of the enhancement method and the domain adaptation for the imbalance of categories in time series data.
MECEM framework
Letting Go of Self-Domain Awareness: Multi-Source Domain-Adversarial Generalization via Dynamic Domain-Weighted Contrastive Transfer Learning
European Conference on Artificial Intelligence 2023
By breaking away from the limitations of traditional self-domain perception, stronger multi-source domain generalization capabilities are achieved through dynamic domain-weighted contrastive learning.
MsCtrl framework