Shuai Lü 0001

dblp:27/10828-1 · also Shuai Lu 0001, Shuai Lv 0001 · DBLP profile ↗
← Back
11ranked-venue papers in the field
2as first author
11since 2021 · last 2027
0000-0002-8081-4498ORCID · verified

Domains — venue-derived; a paper can count in several

Knowledge Engineering, Semantic Web & Information Systems · 7 (2 first)Information Retrieval & Web Search · 4
YearPublicationVenuePosition
2027 Offline-to-online reinforcement learning with triple-intensity policy constraints
Taihong Zhong, Shuai Lü 0001
Inf. Process. Manag.5
2026 Selective Constraint Learning for Unsupervised Cross-Domain Image Retrieval
abstract
Unsupervised cross-domain image retrieval aims to retrieve semantically consistent images across domains with significant domain gaps, which poses substantial challenges under the absence of annotations in both domains. Existing approaches primarily rely on internally derived supervision signals for representation learning and cross-domain alignment. However, such internally induced supervision tends to impose an upper bound on achievable retrieval performance, as it lacks stable semantic references to support reliable category-level correspondence across domains. To address these limitations, we propose Selective Constraint Learning (SCL), a framework that introduces external semantic guidance as a stable prior for unsupervised cross-domain image retrieval. Leveraging a pre-trained foundation model, SCL constructs a dual-scope constraint bank to capture high-confidence positive and negative semantic relations within and across domains. Based on this, we design a generic constraint loss to jointly facilitate intra-domain compactness and inter-domain alignment. In addition, prototypical geometry regularization is designed to enhance in-domain structural stability through prototype-centered pull-and-push forces. Extensive experiments on multiple benchmarks demonstrate that SCL consistently outperforms state-of-the-art methods.
Wensi Fang, Xiaodan Zhang 0006, Xiaoyu Lian, Qiang Li 0008, Shuai Lü 0001
SIGIR5
2026 TATRC: Triple Actor-Critic Structure with Regularization for better performance
Taihong Zhong, Shuai Han 0005, Zehong Long, Shuai Lü 0001, Junhong Wu
Inf. Process. Manag.5
2026 Influence of Gaussian distribution on performance metrics in continuous reinforcement learning
Ruikai Zhou, Taihong Zhong, Wenbo Zhu 0003, Shuai Han 0005, Shuai Lü 0001
Inf. Process. Manag.5
2025 Bi-classifier with neighborhood aggregation for unsupervised domain adaptation
Shuai Lü 0001, Xinyu Zhang 0028, Jingyao Li 0003, Meng Kang
Inf. Sci.1
2024 Explorer-Actor-Critic: Better actors for deep reinforcement learning
Shuai Han 0005, Shuai Lü 0001
Inf. Sci.5
2023 Entropy regularization methods for parameter space exploration
Shuai Han 0005, Wenbo Zhou 0003, Shuai Lü 0001, Xiaoyu Gong
Inf. Sci.3
2022 Actor-critic with familiarity-based trajectory experience replay
Xiaoyu Gong, Jiayu Yu, Shuai Lü 0001, Hengwei Lu
Inf. Sci.3
2022 Unsupervised domain adaptation via softmax-based prototype construction and adaptation
Jingyao Li 0003, Shuai Lü 0001, Zhanshan Li
Inf. Sci.2
2022 Proximal policy optimization via enhanced exploration efficiency
Shuai Han 0005, Shuai Lü 0001
Inf. Sci.4
2021 Recruitment-imitation mechanism for evolutionary reinforcement learning
Shuai Lü 0001, Shuai Han 0005, Wenbo Zhou 0003
Inf. Sci.1