4 papers
Continual Learning with Global Alignment
Xueying Bai, Jinghuan Shang, Yifan Sun +1
Continual learning aims to sequentially learn new tasks without forgetting previous tasks' knowledge (catastrophic forgetting). One factor that can cause forgetting is the interfer…
Unlocking Dynamic Inter-Client Spatial Dependencies: A Federated Spatio-Temporal Graph Learning Method for Traffic Flow Forecasting
Feng Wang, Tianxiang Chen, Shuyue Wei +4
Spatio-temporal graphs are powerful tools for modeling complex dependencies in traffic time series. However, the distributed nature of real-world traffic data across multiple stake…
Cross-Attention Graph Neural Networks for Inferring Gene Regulatory Networks with Skewed Degree Distribution
Jiaqi Xiong, Nan Yin, Shiyang Liang +5
Inferencing Gene Regulatory Networks (GRNs) from gene expression data is a pivotal challenge in systems biology, and several innovative computational methods have been introduced.…
Does RoBERTa Perform Better than BERT in Continual Learning: An Attention Sink Perspective
Xueying Bai, Yifan Sun, Niranjan Balasubramanian
Continual learning (CL) aims to train models that can sequentially learn new tasks without forgetting previous tasks' knowledge. Although previous works observed that pre-training…