paper-with-me

Papers

Multi-scale Feature Alignment for Continual Learning of Unlabeled Domains

2023-02-02 · Kevin Thandiackal, Luigi Piccinelli, Pushpak Pati, Orcun Goksel

Methods for unsupervised domain adaptation (UDA) help to improve the performance of deep neural networks on unseen domains without any labeled data. Especially in medical disciplines such as histopathology, this is crucial since large datasets with detailed annotations are scarce. While the majority of existing UDA methods focus on the adaptation from a labeled source to a single unlabeled target domain, many real-world applications with a long life cycle involve more than one target domain. Thus, the ability to sequentially adapt to multiple target domains becomes essential. In settings where the data from previously seen domains cannot be stored, e.g., due to data protection regulations, the above becomes a challenging continual learning problem. To this end, we propose to use generative feature-driven image replay in conjunction with a dual-purpose discriminator that not only enables the generation of images with realistic features for replay, but also promotes feature alignment during domain adaptation. We evaluate our approach extensively on a sequence of three histopathological datasets for tissue-type classification, achieving state-of-the-art results. We present detailed ablation experiments studying our proposed method components and demonstrate a possible use-case of our continual UDA method for an unsupervised patch-based segmentation task given high-resolution tissue images.

📄 PDF Abstract BibTeX arXiv:2302.01287

Code (0)

등록된 구현이 없습니다.

Tasks

Continual LearningDomain AdaptationUnsupervised Domain Adaptation

Similar Papers 제목 키워드 기반

GOAL: Geometrically Optimal Alignment for Continual Generalized Category Discovery

2026-02-23 · Jizhou Han, Chenhao Ding, SongLin Dong, Yuhang He 외 arxiv

Continual Generalized Category Discovery (C-GCD) requires identifying novel classes from unlabeled data while retaining knowledge of known classes over time. Existing methods typically update classifier weights dynamical…

Novel Class Discovery

Annotation-Free Class-Incremental Learning

2025-11-24 · Hari Chandana Kuchibhotla, K S Ananth, Vineeth N Balasubramanian arxiv

Despite significant progress in continual learning ranging from architectural novelty to clever strategies for mitigating catastrophic forgetting most existing methods rest on a strong but unrealistic assumption the avai…

class-incremental learningContinual Learning

Scalable Adversarial Online Continual Learning

2022-09-04 · Tanmoy Dam, Mahardhika Pratama, Md Meftahul Ferdaus, Sreenatha Anavatti 외

Adversarial continual learning is effective for continual learning problems because of the presence of feature alignment process generating task-invariant features having low susceptibility to the catastrophic forgetting…

Continual LearningMeta-Learning

Unsupervised Continual Domain Shift Learning with Multi-Prototype Modeling

2025-01-01 · CVPR 2025 1 · Haopeng Sun, Yingwei Zhang, Lumin Xu, Sheng Jin 외

In real-world applications, deep neural networks may encounter constantly changing environments, where the test data originates from continually shifting unlabeled target domains. This problem, known as Unsupervised …

Unsupervised Continual Domain Shift Learning

Divide-and-Conquer for Enhancing Unlabeled Learning, Stability, and Plasticity in Semi-supervised Continual Learning

2025-08-07 · Yue Duan, Taicai Chen, Lei Qi, Yinghuan Shi arxiv

Semi-supervised continual learning (SSCL) seeks to leverage both labeled and unlabeled data in a sequential learning setup, aiming to reduce annotation costs while managing continual data arrival. SSCL introduces complex…

Continual Learning