Efficient Federated-Learning Model Debugging

Li, Anran; Zhang, Lan; Wang, Junhao; Tan, Juntao; Han, Feng; Qin, Yaxuan; Freris, Nikolaos M.; Li, Xiangyang

doi:10.1109/icde51399.2021.00039

Cited by 30 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Table 2, both strategies select 3 identical clients (2,6,18). A comparison of the other 2 clients selected by the two strategies shows that our selection of clients 15 and 16 contains data samples from all categories, while the clients selected by DICE (i.e., 0 and 1) omit these rare categories (i.e., 7, 8, and 9).…”

Section: ) Rsmentioning

confidence: 97%

“…However, the ensuing data privacy issues pose new challenges for model developers [1]. As a result, scholars have proposed Federated Learning (FL) -a distributed computational training method designed to protect user privacy [2]. FL takes full advantage of the abundant data resources and computational power on each device, with the server as the coordination center.…”

Section: Introductionmentioning

confidence: 99%

“…We compared the selection outcomes of our approach with the best-performing baseline DICE in the selection of 5 clients out of 20 from CIFAR10-D6. Please note that D6 is an extreme case with a highly imbalanced global data distribution (refer to Figure8a).As shown in Table2, both strategies select 3 identical clients(2,6,18). A comparison of the other 2 clients selected by the two strategies shows that our selection of clients 15 and 16 contains data samples from all categories, while the clients selected by DICE (i.e., 0 and 1) omit these rare categories (i.e., 7, 8, and 9).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Pretraining Client Selection Algorithm Based on a Data Distribution Evaluation Model in Federated Learning

Xu,

Liu,

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Federated Learning (FL) allows task initiators (servers) to utilize data from task participants (clients) to train machine learning models while protecting data privacy. However, in the FL system, when the client data are non-independently identically distributed (Non-IID), appropriate metrics are chosen to accurately evaluate the quality of the client data, accordingly to select a reasonable subset of clients, and thus ensure the accuracy of the FL aggregation model. In this paper, based on the experimental results, a data distribution evaluation model is proposed, which is based on two metrics: the volume of client data and its increment and the balance of global client data. This data distribution evaluation model enables more accurate evaluation of clients with Non-IID characteristics. Based on this evaluation model, this paper further proposes an FL client subset selection algorithm. This algorithm accurately evaluates the data value of each client, enabling the server to select the most valuable subset of clients before FL training, thus improving the accuracy of the federated learning aggregation model in scenarios with Non-IID client data. When training the FL aggregation model using the proposed method on datasets composed of CIFAR-10, Fashion-MNIST, and DEAP distributions, compared to the optimal baseline, the average precision scores increased by 5.99%, 4.79%, and 4.29% respectively. The improvement in accuracy is more pronounced in scenarios with Non-IID data, such as in the DEAP dataset distribution with the highest Non-IID degree, where the accuracy increased by 5.30% compared to the optimal baseline.

show abstract

Section: ) Rsmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Pretraining Client Selection Algorithm Based on a Data Distribution Evaluation Model in Federated Learning

Xu,

Liu,

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, it limits machine learning's capability to deal with applications where data has been isolated across different organizations and data privacy has been emphasized [1], [2], e.g., user's private pictures and videos [3]- [5] captured by mobile phone as well as social relationships [6] should not be leaked during model training. Federated learning (FL) [7]- [10] is one emerging technology, which enables multiple parties to collaboratively train a machine learning model by iteratively exchanging model parameters between these parties and a centralized server, meanwhile keeping their datasets private. There are three types of FL methods according to the distribution of data, which are horizontal federated learning (HFL) [7], [11], vertical federated learning (VFL) [12], [13], and federated transfer learning (FTL) [14] respectively.…”

Section: Introductionmentioning

confidence: 99%

Residue-based Label Protection Mechanisms in Vertical Logistic Regression

Tan¹,

Zhang²,

Liu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Federated learning (FL) enables distributed participants to collaboratively learn a global model without revealing their private data to each other. Recently, vertical FL, where the participants hold the same set of samples but with different features, has received increased attention. This paper first presents one label inference attack method to investigate the potential privacy leakages of the vertical logistic regression model. Specifically, we discover that the attacker can utilize the residue variables, which are calculated by solving the system of linear equations constructed by local dataset and the received decrypted gradients, to infer the privately owned labels. To deal with this, we then propose three protection mechanisms, e.g., additive noise mechanism, multiplicative noise mechanism, and hybrid mechanism which leverages local differential privacy and homomorphic encryption techniques, to prevent the attack and improve the robustness of the vertical logistic regression model. Experimental results show that both the additive noise mechanism and the multiplicative noise mechanism can achieve efficient label protection with only a slight drop in model testing accuracy, furthermore, the hybrid mechanism can achieve label protection without any testing accuracy degradation, which demonstrates the effectiveness and efficiency of our protection techniques.

show abstract

“…mobile edge computing (MEC) scenarios due to the limitation of data privacy and decentralization. As a popular privacy-preserving and distributed AI paradigm, Federated learning (FL) [21,32,[34][35][36]38] is promising to become the infrastructure for AI applications in IoT [3,12,16,24,29,38] and MEC [11,18,19,33,34] scenarios. By dispatching a global model to multiple clients for local training and then aggregating the trained models, FL realizes distributed model training without data sharing, which also greatly reduces the risk of data privacy leakage.…”

Section: Introductionmentioning

confidence: 99%

FedCross: Towards Accurate Federated Learning via Multi-Model Cross Aggregation

Hu¹,

Zhou²,

Yue³

et al. 2022

Preprint

View full text Add to dashboard Cite

Due to the remarkable performance in preserving data privacy for decentralized data scenarios, Federated Learning (FL) has been considered as a promising distributed machine learning paradigm to deal with data silos problems. Typically, conventional FL approaches adopts a one-to-multi training scheme, where the cloud server keeps only one single global model for all the involved clients for the purpose of model aggregation. However, this scheme suffers from inferior classification performance, since only one global model cannot always accommodate all the incompatible convergence directions of local models, resulting in low convergence rate and classification accuracy. To address this issue, this paper presents an efficient FL framework named FedCross, which adopts a novel multi-to-multi FL training scheme based on our proposed similaritybased multi-model cross aggregation method. Unlike traditional FL methods, in each round of FL training, FedCross uses a small set of distinct intermediate models to conduct weighted fusion under the guidance of model similarities. In this way, the intermediate models used by FedCross can sufficiently respect the convergence characteristics of clients, thus leading to much fewer conflicts in tuning the convergence directions of clients. Finally, in the deployment stage, FedCross forms a global model for all the clients by performing the federated averaging on the trained immediate models. Experimental results on three well-known datasets demonstrate that, compared with state-of-the-art FL approaches, FedCross can significantly improve FL accuracy in both IID and non-IID scenarios without causing additional communication overhead.

show abstract

Efficient Federated-Learning Model Debugging

Cited by 30 publications

References 14 publications

Pretraining Client Selection Algorithm Based on a Data Distribution Evaluation Model in Federated Learning

Pretraining Client Selection Algorithm Based on a Data Distribution Evaluation Model in Federated Learning

Residue-based Label Protection Mechanisms in Vertical Logistic Regression

FedCross: Towards Accurate Federated Learning via Multi-Model Cross Aggregation

Contact Info

Product

Resources

About