J. Cho
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
J. Cho has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
J. Shin
Gwinnett Technical College Intel Corporation Prairie View A&M University Ohio State University
1 shared publication- Jae-Jin Lee
Dept. of Electr. & Comput. Eng., Univ. of Alabama, Tuscaloosa, AL, USA
1 shared publication - Jinho Cha
Gwinnett Technical College
1 shared publication
Latest Publications
- Open Access
FOSSIL: Regret-minimizing weighting for robust learning under imbalance and small data
Imbalanced and small data regimes are pervasive in domains such as rare disease imaging, genomics, and disaster response, where labeled samples are scarce and naive augmentation often introduces artifacts. Existing solutions such as oversampling, focal loss, or meta-weighting address isolated aspects of this challenge but remain fragile or complex. We introduce FOSSIL (Flexible Optimization via Sample Sensitive Importance Learning), a unified weighting framework that seamlessly integrates class imbalance correction, difficulty-aware curricula, augmentation penalties, and warmup dynamics into a single interpretable formula. Unlike prior heuristics, the proposed framework provides regret-based theoretical guarantees and achieves consistent empirical gains over ERM, curriculum, and meta-weighting baselines on synthetic and real-world datasets, while requiring no architectural changes.
Open AccessFOSSIL: Regret-minimizing weighting for robust learning under imbalance and small data
Imbalanced and small data regimes are pervasive in domains such as rare disease imaging, genomics, and disaster response, where labeled samples are scarce and naive augmentation often introduces artifacts. Existing solutions such as oversampling, focal loss, or meta-weighting address isolated aspects of this challenge but remain fragile or complex. We introduce FOSSIL (Flexible Optimization via Sample Sensitive Importance Learning), a unified weighting framework that seamlessly integrates class imbalance correction, difficulty-aware curricula, augmentation penalties, and warmup dynamics into a single interpretable formula. Unlike prior heuristics, the proposed framework provides regret-based theoretical guarantees and achieves consistent empirical gains over ERM, curriculum, and meta-weighting baselines on synthetic and real-world datasets, while requiring no architectural changes.