{"type":"ScholarlyArticle","title":"Towards Robust Incremental Learning Under Ambiguous Supervision","authors":[{"name":"Rui Wang","slug":"rui-wang","given":"Rui","family":"Wang","affiliation":""},{"name":"Mingxuan Xia","slug":"mingxuan-xia","given":"Mingxuan","family":"Xia","affiliation":""},{"name":"Haobo Wang","slug":"haobo-wang","given":"Haobo","family":"Wang","affiliation":""},{"name":"Lei Feng","slug":"lei-feng","given":"Lei","family":"Feng","affiliation":""},{"name":"Junbo Zhao","slug":"junbo-zhao","given":"Junbo","family":"Zhao","affiliation":""},{"name":"Gang Chen","slug":"gang-chen","given":"Gang","family":"Chen","affiliation":""},{"name":"Chang Yao","slug":"chang-yao","given":"Chang","family":"Yao","affiliation":""}],"abstract":"Traditional Incremental Learning (IL) targets to handle sequential fully-supervised learning problems where novel classes emerge from time to time. However, due to inherent annotation uncertainty and ambiguity, collecting high-quality annotated data in a dynamic learning system can be extremely expensive. To mitigate this problem, we propose a novel weakly-supervised learning paradigm called Incremental Partial Label Learning (IPLL), where the sequentially arrived data relate to a set of candidate labels rather than the ground truth. Technically, we develop the Prototype-Guided Disambiguation and Replay Algorithm (PGDR) which leverages the class prototypes as a proxy to mitigate two intertwined challenges in IPLL, i.e., label ambiguity and catastrophic forgetting. To handle the former, PGDR encapsulates a momentum-based pseudo-labeling algorithm along with prototype-guided initialization, resulting in a balanced perception of classes. To alleviate forgetting, we develop a memory replay technique that collects well-disambiguated samples while maintaining representativeness and diversity. By jointly distilling knowledge from curated memory data, our framework exhibits a great disambiguation ability for samples of new tasks and achieves less forgetting of knowledge. Extensive experiments demonstrate that PGDR achieves superior performance over the baselines in the IPLL task.","doi":"10.24963/ijcai.2025/712","url":"https://ijcai.org/proceedings/2025/712","pdf":"https://ijcai.org/proceedings/2025/0712.pdf","markdown":"https://ijcai.org/proceedings/2025/712.md","bibtex":"https://ijcai.org/proceedings/2025/bibtex/712","volume":{"slug":"2025","title":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2025","url":"https://ijcai.org/proceedings/2025/","year":2025,"location":"Montreal, Canada","dates":"16-22 August 2025","editors":"James Kwok","isbn":"978-1-956792-06-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2025"},"section":"Machine Learning","track":"Main Track","number":712,"pages":{"first":6397,"last":6405},"keywords":["Machine Learning: ML: Weakly supervised learning","Machine Learning: ML: Semi-supervised learning"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}