Privacy-Preserving Federated Multi-Task Linear Regression: A One-Shot Linear Mixing Approach Inspired By Graph Regularization

Lee, Harlin; Bertozzi, Andrea L.; Kovačević, Jelena; Chi, Yuejie

doi:10.1109/icassp43922.2022.9746007

Cited by 4 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In practical scenarios, the data is often dispersed across different silos, requiring a federated approach to estimate causal relationships due to privacy constraints [7][8][9][10]. However, differences in data feature dimensions and sample sizes across silos can introduce local biases when estimating causal effects, such as variations in medical records across different hospitals for the same patient.…”

Section: Introductionmentioning

confidence: 99%

Geography of Technology Transfer in China

Liu

2023

East China Normal University Scientific Reports

View full text Add to dashboard Cite

Graphs can model complex relationships between objects, enabling a myriad of Web applications such as online page/article classification and social recommendation. While graph neural networks (GNNs) have emerged as a powerful tool for graph representation learning, in an end-to-end supervised setting, their performance heavily relies on a large amount of task-specific supervision. To reduce labeling requirement, the "pre-train, fine-tune" and "pre-train, prompt" paradigms have become increasingly common. In particular, prompting is a popular alternative to fine-tuning in natural language processing, which is designed to narrow the gap between pre-training and downstream objectives in a task-specific manner. However, existing study of prompting on graphs is still limited, lacking a universal treatment to appeal to different downstream tasks. In this paper, we propose GraphPrompt, a novel pre-training and prompting framework on graphs. GraphPrompt not only unifies pre-training and downstream tasks into a common task template, but also employs a learnable prompt to assist a downstream task in locating the most relevant knowledge from the pre-trained model in a task-specific manner. Finally, we conduct extensive experiments on five public datasets to evaluate and analyze GraphPrompt. CCS CONCEPTS• Computing methodologies → Learning latent representations; • Information systems → Data mining.

show abstract

Section: Introductionmentioning

confidence: 99%

Geography of Technology Transfer in China

Liu

2023

East China Normal University Scientific Reports

View full text Add to dashboard Cite

show abstract

“…Different from the bilevel optimization, some works [26], [27] train the local models and the GNN model sequentially with separate objective functions. For example, clients in [27] train their local models with different local tasks. Then, the server trains a GNN model to fuse multitask local estimates.…”

Section: A Centralized Fedgnnsmentioning

confidence: 99%

“…FedGNNs have been applied in KG completion [87], [89], [90], [92] and recommendation system tasks [73], [83], [84] with privacy protection by considering one KG or one user as one client. Besides, income prediction [27], malicious transaction detection [44], [45], network anomaly detection [68],…”

Section: A Applicationsmentioning

confidence: 99%

“…2) Simulated Graph Based on Clients' Similarities: Some datasets have no natural graph topology. The interclient graph can be calculated based on the clients' information, such as clients' embedding [23], [27], [27], [28], [30], [37]. 3) Synthetic Graph: A fully connected graph [17], [42], ring connected graph [32], or just a randomly generated graph can be used as the interclient graph [34], [35], [40].…”

Section: Fedgnn Experimental Evaluation Designmentioning

confidence: 99%

“…3) Label Distribution Skew Partition: Different clients have different subsets of label classes. It can be achieved by making one client maintain data only from one class or a subset of classes[27],[30]. 4) Natural Identity Partition: Different clients have different characteristics.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Federated Graph Neural Networks: Overview, Techniques, and Challenges

Liu,

Xing,

Deng

et al. 2024

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

Graph neural networks (GNNs) have attracted extensive research attention in recent years due to their capability to progress with graph data and have been widely used in practical applications. As societies become increasingly concerned with the need for data privacy protection, GNNs face the need to adapt to this new normal. Besides, as clients in federated learning (FL) may have relationships, more powerful tools are required to utilize such implicit information to boost performance. This has led to the rapid development of the emerging research field of federated GNNs (FedGNNs). This promising interdisciplinary field is highly challenging for interested researchers to grasp. The lack of an insightful survey on this topic further exacerbates the entry difficulty. In this article, we bridge this gap by offering a comprehensive survey of this emerging field. We propose a 2-D taxonomy of the FedGNN literature: 1) the main taxonomy provides a clear perspective on the integration of GNNs and FL by analyzing how GNNs enhance FL training as well as how FL assists GNN training and 2) the auxiliary taxonomy provides a view on how FedGNNs deal with heterogeneity across FL clients. Through discussions of key ideas, challenges, and limitations of existing works, we envision future research directions that can help build more robust, explainable, efficient, fair, inductive, and comprehensive FedGNNs.

show abstract