Paper Detail
Quyen Tran, Hai Nguyen, Hoang Phan, Quan Dao, Linh Ngo, Khoat Than, Dinh Phung, Dimitris Metaxas, Trung Le
In online incremental learning, data continuously arrives with substantial distributional shifts, creating a significant challenge because previous samples have limited replay value when learning a new task. Prior research has typically relied on either a single adaptive centroid or multiple fixed centroids to represent each class in the latent space. However, such methods struggle when class data streams are inherently multimodal and require continual centroid updates. To overcome this, we introduce an online Mixture Model learning framework grounded in Optimal Transport theory (MMOT), where centroids evolve incrementally with new data. This approach offers two main advantages: (i) it provides a more precise characterization of complex data streams, and (ii) it enables improved class similarity estimation for unseen samples during inference through MMOT-derived centroids. Furthermore, to strengthen representation learning and mitigate catastrophic forgetting, we design a Dynamic Preservation strategy that regulates the latent space and maintains class separability over time. Experimental evaluations on benchmark datasets confirm the superior effectiveness of our proposed method.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{tran2026optimal,
title = {An Optimal Transport-driven Approach for Cultivating Latent Space in Online Incremental Learning},
author = {Quyen Tran and Hai Nguyen and Hoang Phan and Quan Dao and Linh Ngo and Khoat Than and Dinh Phung and Dimitris Metaxas and Trung Le},
year = {2026},
abstract = {In online incremental learning, data continuously arrives with substantial distributional shifts, creating a significant challenge because previous samples have limited replay value when learning a new task. Prior research has typically relied on either a single adaptive centroid or multiple fixed centroids to represent each class in the latent space. However, such methods struggle when class data streams are inherently multimodal and require continual centroid updates. To overcome this, we intr},
url = {https://huggingface.co/papers/2211.16780},
keywords = {online incremental learning, distributional shifts, adaptive centroid, fixed centroids, latent space, optimal transport theory, mixture model, centroid evolution, class similarity estimation, catastrophic forgetting, representation learning, Dynamic Preservation strategy, class separability, huggingface daily},
eprint = {2211.16780},
archiveprefix = {arXiv},
}
{}