The Experts below are selected from a list of 4209 Experts worldwide ranked by ideXlab platform
Hithnawi Anwar - One of the best experts on this subject based on the ideXlab platform.
-
RoFL: Attestable Robustness for Secure Federated Learning
2021Co-Authors: Burkhalter Lukas, Lycklama Hidde, Viand Alexander, Küchler Nicolas, Hithnawi AnwarAbstract:Federated Learning is an emerging decentralized machine learning paradigm that allows a large number of Clients to train a joint model without the need to share their private data. Participants instead only share ephemeral updates necessary to train the model. To ensure the confidentiality of the Client updates, Federated Learning systems employ secure aggregation; Clients encrypt their gradient updates, and only the aggregated model is revealed to the server. Achieving this level of data protection, however, presents new challenges to the robustness of Federated Learning, i.e., the ability to tolerate failures and attacks. Unfortunately, in this setting, a Malicious Client can now easily exert influence on the model behavior without being detected. As Federated Learning is being deployed in practice in a range of sensitive applications, its robustness is growing in importance. In this paper, we take a step towards understanding and improving the robustness of secure Federated Learning. We start this paper with a systematic study that evaluates and analyzes existing attack vectors and discusses potential defenses and assesses their effectiveness. We then present RoFL, a secure Federated Learning system that improves robustness against Malicious Clients through input checks on the encrypted model updates. RoFL extends Federated Learning's secure aggregation protocol to allow expressing a variety of properties and constraints on model updates using zero-knowledge proofs. To enable RoFL to scale to typical Federated Learning settings, we introduce several ML and cryptographic optimizations specific to Federated Learning. We implement and evaluate a prototype of RoFL and show that realistic ML models can be trained in a reasonable time while improving robustness.Comment: 20 pages, 15 figures. Updated last name of one author to improve indexabilit
Burkhalter Lukas - One of the best experts on this subject based on the ideXlab platform.
-
RoFL: Attestable Robustness for Secure Federated Learning
2021Co-Authors: Burkhalter Lukas, Lycklama Hidde, Viand Alexander, Küchler Nicolas, Hithnawi AnwarAbstract:Federated Learning is an emerging decentralized machine learning paradigm that allows a large number of Clients to train a joint model without the need to share their private data. Participants instead only share ephemeral updates necessary to train the model. To ensure the confidentiality of the Client updates, Federated Learning systems employ secure aggregation; Clients encrypt their gradient updates, and only the aggregated model is revealed to the server. Achieving this level of data protection, however, presents new challenges to the robustness of Federated Learning, i.e., the ability to tolerate failures and attacks. Unfortunately, in this setting, a Malicious Client can now easily exert influence on the model behavior without being detected. As Federated Learning is being deployed in practice in a range of sensitive applications, its robustness is growing in importance. In this paper, we take a step towards understanding and improving the robustness of secure Federated Learning. We start this paper with a systematic study that evaluates and analyzes existing attack vectors and discusses potential defenses and assesses their effectiveness. We then present RoFL, a secure Federated Learning system that improves robustness against Malicious Clients through input checks on the encrypted model updates. RoFL extends Federated Learning's secure aggregation protocol to allow expressing a variety of properties and constraints on model updates using zero-knowledge proofs. To enable RoFL to scale to typical Federated Learning settings, we introduce several ML and cryptographic optimizations specific to Federated Learning. We implement and evaluate a prototype of RoFL and show that realistic ML models can be trained in a reasonable time while improving robustness.Comment: 20 pages, 15 figures. Updated last name of one author to improve indexabilit
Asokan N. - One of the best experts on this subject based on the ideXlab platform.
-
WAFFLE: Watermarking in Federated Learning
2021Co-Authors: Atli, Buse Gul, Xia Yuxi, Marchal Samuel, Asokan N.Abstract:Federated learning is a distributed learning technique where machine learning models are trained on Client devices in which the local training data resides. The training is coordinated via a central server which is, typically, controlled by the intended owner of the resulting model. By avoiding the need to transport the training data to the central server, federated learning improves privacy and efficiency. But it raises the risk of model theft by Clients because the resulting model is available on every Client device. Even if the application software used for local training may attempt to prevent direct access to the model, a Malicious Client may bypass any such restrictions by reverse engineering the application software. Watermarking is a well-known deterrence method against model theft by providing the means for model owners to demonstrate ownership of their models. Several recent deep neural network (DNN) watermarking techniques use backdooring: training the models with additional mislabeled data. Backdooring requires full access to the training data and control of the training process. This is feasible when a single party trains the model in a centralized manner, but not in a federated learning setting where the training process and training data are distributed among several Client devices. In this paper, we present WAFFLE, the first approach to watermark DNN models trained using federated learning. It introduces a retraining step at the server after each aggregation of local models into the global model. We show that WAFFLE efficiently embeds a resilient watermark into models incurring only negligible degradation in test accuracy (-0.17%), and does not require access to training data. We also introduce a novel technique to generate the backdoor used as a watermark. It outperforms prior techniques, imposing no communication, and low computational (+3.2%) overhead.Comment: Will appear in the proceedings of SRDS 2021; 14 pages, 11 figures, 10 table
Roedig Utz - One of the best experts on this subject based on the ideXlab platform.
-
The benefits of deceit: a Malicious Client in a 5G cellular network
'Institute of Electrical and Electronics Engineers (IEEE)', 2019Co-Authors: Quinlan, Jason J., Roedig UtzAbstract:As we advance towards Smart Cities, autonomous vehicles and the avalanche of IoT devices proposed for the future, we need to give careful consideration to how easily compromised nodes/devices can impact network state. Current proposals for autonomous smart devices typically use cellular networks as the backhaul or final hop. These devices will leverage existing trust-based Client-side channel metrics, such as channel quality indicator (CQI), when the base-station determines scheduling decisions. In this short paper, we investigate the scheduling impact of a Malicious device when it changes its channel metrics, so as to improve its download rate or even to negate the download rate of others. We utilise real-time 4K ultra-high definition video delivery as an example of high throughput demand application and compare the delivery rates of multiple devices in an open-source 5G simulated NS-3 network. Our results illustrate that when a Malicious Client deceives the scheduler, the other Clients in the network have a noticeable decrease in both viewable quality and underlying delivery rate (25% decrease in the average video quality across the non Malicious Clients)
Küchler Nicolas - One of the best experts on this subject based on the ideXlab platform.
-
RoFL: Attestable Robustness for Secure Federated Learning
2021Co-Authors: Burkhalter Lukas, Lycklama Hidde, Viand Alexander, Küchler Nicolas, Hithnawi AnwarAbstract:Federated Learning is an emerging decentralized machine learning paradigm that allows a large number of Clients to train a joint model without the need to share their private data. Participants instead only share ephemeral updates necessary to train the model. To ensure the confidentiality of the Client updates, Federated Learning systems employ secure aggregation; Clients encrypt their gradient updates, and only the aggregated model is revealed to the server. Achieving this level of data protection, however, presents new challenges to the robustness of Federated Learning, i.e., the ability to tolerate failures and attacks. Unfortunately, in this setting, a Malicious Client can now easily exert influence on the model behavior without being detected. As Federated Learning is being deployed in practice in a range of sensitive applications, its robustness is growing in importance. In this paper, we take a step towards understanding and improving the robustness of secure Federated Learning. We start this paper with a systematic study that evaluates and analyzes existing attack vectors and discusses potential defenses and assesses their effectiveness. We then present RoFL, a secure Federated Learning system that improves robustness against Malicious Clients through input checks on the encrypted model updates. RoFL extends Federated Learning's secure aggregation protocol to allow expressing a variety of properties and constraints on model updates using zero-knowledge proofs. To enable RoFL to scale to typical Federated Learning settings, we introduce several ML and cryptographic optimizations specific to Federated Learning. We implement and evaluate a prototype of RoFL and show that realistic ML models can be trained in a reasonable time while improving robustness.Comment: 20 pages, 15 figures. Updated last name of one author to improve indexabilit