3 May 2022 in ACM Transactions on Intelligent Systems and Technology
Data privacy has become an increasingly important issue in Machine Learning (ML) , where many approaches have been developed to tackle this challenge, e.g., cryptography ( Homomorphic Encryption (HE) , Differential Privacy (DP) ) and collaborative training (Secure Multi-Party Computation (MPC) , Distributed Learning, and Federated Learning (FL) ). These techniques have a particular focus on data encryption or secure local computation. They transfer the intermediate information to the third party to compute the final result. Gradient exchanging is commonly considered to be a secure way of training a robust model collaboratively in Deep Learning (DL) . However, recent researches have demonstrated that sensitive information can be recovered from the shared gradient. Generative Adversarial Network (GAN) , in particular, has shown to be effective in recovering such information. However, GAN based techniques require additional information, such as class labels that are generally unavailable for privacy-preserved learning. In this article, we show that, in the FL system, image-based privacy data can be easily recovered in full from the shared gradient only via our proposed Generative Regression Neural Network (GRNN) . We formulate the attack to be a regression problem and optimize two branches of the generative model by minimizing the distance between gradients. We evaluate our method on several image classification tasks. The results illustrate that our proposed GRNN outperforms state-of-the-art methods with better stability, stronger robustness, and higher accuracy. It also has no convergence requirement to the global FL model. Moreover, we demonstrate information leakage using face re-identification. Some defense strategies are also discussed in this work.
25 May 2024 in ICCK Transactions on Intelligent Systematics
To predict future trends based on the data from sensors is an important technology for many applications, such as the Internet of Things, smart cities, etc. Based on the predicted results, further decisions and system controls can be made. Raw sensor data sets are often complex non-linear data with noise, which results in the difficulty of accurate prediction. This paper proposes a distributed deep prediction network based on a covariance intersection (CI) fusion algorithm in which the deep learning networks, such as long short-term memory networks (LSTM) and gated recurrent unit networks (GRU) are fused by CI fusion algorithm to effectively improve the performance of prediction. Moreover, the variance is obtained to evaluate the prediction results. The model is validated on the real weather dataset in Beijing. The experiments show that LSTM and GRU have their pros and cons for different data, CI fusion can improve the accuracy of the final predictions, and the entire framework has robust prediction results with a reasonable estimated variance.
25 September 2026 in SSRN Electronic Journal
3D Gaussian Splatting (3DGS) enables high-quality real-time novel-view synthesis but incurs substantial storage overhead due to the large number of Gaussian primitives required for complex scenes. Existing compression methods either rely on expensive per-scene retraining or apply uniform post-training strategies that ignore the heterogeneous importance of individual Gaussians. We present FA-Compress, a training-free post-processing framework that exploits this heterogeneity: primitives in smooth regions tolerate substantially more compression than those encoding high-frequency structures. FA-Compress estimates per-Gaussian frequency characteristics directly from model-internal attributes and performs frequency-guided SH truncation, opacity pruning, and adaptive quantization without retraining or rendering-based optimization. On NeRF Synthetic, Mip-NeRF 360, and Tanks & Temples, FA-Compress consistently achieves higher rendering quality than existing training-free methods and competitive quality against retraining-based approaches, completing in seconds while remaining fully compatible with the standard 3DGS rendering pipeline.