No abstract

While Federated learning (FL) is attractive for pulling privacy-preserving distributed training data, the credibility of participating clients and non-inspectable data pose new security threats, of which poisoning attacks are particularly rampant and hard to defend without compromising privacy, performance or other desirable properties of FL. To tackle this problem, we propose a self-purified FL (SPFL) method that enables benign clients to exploit trusted historical features of locally purified model to supervise the training of aggregated model in each iteration. The purification is performed by an attention-guided self-knowledge distillation where the teacher and student models are optimized locally for task loss, distillation loss and attention-based loss simultaneously. SPFL imposes no restriction on the communication protocol and aggregator at the server. It can work in tandem with any existing secure aggregation algorithms and protocols for augmented security and privacy guarantee. We experimentally demonstrate that SPFL outperforms state-of-the-art FL defenses against various poisoning attacks. The attack success rate of SPFL trained model is at most 3% above that of a clean model, even if the poisoning attack is launched in every iteration with all but one malicious clients in the system. Meantime, it improves the model quality on normal inputs compared to FedAvg, either under attack or in the absence of an attack.

Backdoor attacks have emerged as an urgent threat to Deep Neural Networks (DNNs), where victim DNNs are furtively implanted with malicious neurons that could be triggered by the adversary. To defend against backdoor attacks, many works establish a staged pipeline to remove backdoors from victim DNNs: inspecting, locating, and erasing. However, in a scenario where a few clean data can be accessible, such pipeline is fragile and cannot erase backdoors completely without sacrificing model accuracy. To address this issue, in this paper, we propose a novel data-free holistic backdoor erasing (DHBE) framework. Instead of the staged pipeline, the DHBE treats the backdoor erasing task as a unified adversarial procedure, which seeks equilibrium between two different competing processes: distillation and backdoor regularization. In distillation, the backdoored DNN is distilled into a proxy model, transferring its knowledge about clean data, yet backdoors are simultaneously transferred. In backdoor regularization, the proxy model is holistically regularized to prevent from infecting any possible backdoor transferred from distillation. These two processes jointly proceed with data-free adversarial optimization until a clean, high-accuracy proxy model is obtained. With the novel adversarial design, our framework demonstrates its superiority in three aspects: 1) minimal detriment to model accuracy, 2) high tolerance for hyperparameters, and 3) no demand for clean data. Extensive experiments on various backdoor attacks and datasets are performed to verify the effectiveness of the proposed framework. Code is available at https://github.com/yanzhicong/DHBE

Object detection in the 2D domain is well developed owing to the wide application of CMOS image sensors and the great success of deep learning technologies in recent years. However, under circumstances such as autonomous driving, the variation of weather conditions and light conditions makes it impossible to perform reliable detection using regular 2D image sensors. 3D data generated by a Lidar or Radar is more robust to such environments, hence serving as an essential complement to 2D data in such scenarios. Well-established anchor-based detectors in the 2D domain suffer from time-consuming anchor configuration and cannot be exploited directly to process 3D data. This paper proposes an anchor-free network that encodes the raw point cloud into a hierarchical pillar representation to locate objects. Without predefined anchors and NMS postprocessing, our method directly predicts the center points and box properties to accomplish the detection task efficiently. In addition, a PCA-based initialization for the convolutional kernel is proposed to accelerate the training process. Experiments are implemented on the KITTI benchmark, and our method can achieve competitive performance with other anchor-based methods. Comprehensive ablation studies further verify the validity and rationality of each part of the proposed method.