Disentangled Representation Learning for Multiple Attributes Preserving Face Deidentification.

Face is one of the most attractive sensitive information in visual shared data. It is an urgent task to design an effective face deidentification method to achieve a balance between facial privacy protection and data utilities when sharing data. Most of the previous methods for face deidentification rely on attribute supervision to preserve a certain kind of identity-independent utility but lose the other identity-independent data utilities. In this article, we mainly propose a novel disentangled representation learning architecture for multiple attributes preserving face deidentification called replacing and restoring variational autoencoders (R2VAEs). The R2VAEs disentangle the identity-related factors and the identity-independent factors so that the identity-related information can be obfuscated, while they do not change the identity-independent attribute information. Moreover, to improve the details of the facial region and make the deidentified face blends into the image scene seamlessly, the image inpainting network is employed to fill in the original facial region by using the deidentified face as a priori. Experimental results demonstrate that the proposed method effectively deidentifies face while maximizing the preservation of the identity-independent information, which ensures the semantic integrity and visual quality of shared images.

A Multifactorial Optimization Framework Based on Adaptive Intertask Coordinate System.

The searching ability of the population-based search algorithms strongly relies on the coordinate system on which they are implemented. However, the widely used coordinate systems in the existing multifactorial optimization (MFO) algorithms are still fixed and might not be suitable for various function landscapes with differential modalities, rotations, and dimensions; thus, the intertask knowledge transfer might not be efficient. Therefore, this article proposes a novel intertask knowledge transfer strategy for MFOs implemented upon an active coordinate system that is established on a common subspace of two search spaces. The proper coordinate system might identify some common modality in a proper subspace to some extent. In this article, to seek the intermediate subspace, we innovatively introduce the geodesic flow that starts from a subspace, reaching another subspace in unit time. A low-dimension intermediate subspace is drawn from a uniform distribution defined on the geodesic flow, and the corresponding coordinate system is given. The intertask trial generation method is applied to the individuals by first projecting them on the low-dimension subspace, which reveals the important invariant features of the multiple function landscapes. Since intermediate subspace is generated from the major eigenvectors of tasks' spaces, this model turns out to be intrinsically regularized by neglecting the minor and small eigenvalues. Therefore, the transfer strategy can alleviate the influence of noise led by redundant dimensions. The proposed method exhibits promising performance in the experiments.

Deep Neural Message Passing With Hierarchical Layer Aggregation and Neighbor Normalization.

As a unified framework for graph neural networks, message passing-based neural network (MPNN) has attracted a lot of research interest and has been shown successfully in a number of domains in recent years. However, because of over-smoothing and vanishing gradients, deep MPNNs are still difficult to train. To alleviate these issues, we first introduce a deep hierarchical layer aggregation (DHLA) strategy, which utilizes a block-based layer aggregation to aggregate representations from different layers and transfers the output of the previous block to the subsequent block, so that deeper MPNNs can be easily trained. Additionally, to stabilize the training process, we also develop a novel normalization strategy, neighbor normalization (NeighborNorm), which normalizes the neighbor of each node to further address the training issue in deep MPNNs. Our analysis reveals that NeighborNorm can smooth the gradient of the loss function, i.e., adding NeighborNorm makes the optimization landscape much easier to navigate. Experimental results on two typical graph pattern-recognition tasks, including node classification and graph classification, demonstrate the necessity and effectiveness of the proposed strategies for graph message-passing neural networks.

Locality preserving dense graph convolutional networks with graph context-aware node representations.

Graph convolutional networks (GCNs) have been widely used for representation learning on graph data, which can capture structural patterns on a graph via specifically designed convolution and readout operations. In many graph classification applications, GCN-based approaches have outperformed traditional methods. However, most of the existing GCNs are inefficient to preserve local information of graphs - a limitation that is especially problematic for graph classification. In this work, we propose a locality-preserving dense GCN with graph context-aware node representations. Specifically, our proposed model incorporates a local node feature reconstruction module to preserve initial node features into node representations, which is realized via a simple but effective encoder-decoder mechanism. To capture local structural patterns in neighborhoods representing different ranges of locality, dense connectivity is introduced to connect each convolutional layer and its corresponding readout with all previous convolutional layers. To enhance node representativeness, the output of each convolutional layer is concatenated with the output of the previous layer's readout to form a global context-aware node representation. In addition, a self-attention module is introduced to aggregate layer-wise representations to form the final graph-level representation. Experiments on benchmark datasets demonstrate the superiority of the proposed model over state-of-the-art methods in terms of classification accuracy.

MGAT: Multi-view Graph Attention Networks.

Multi-view graph embedding is aimed at learning low-dimensional representations of nodes that capture various relationships in a multi-view network, where each view represents a type of relationship among nodes. Multitudes of existing graph embedding approaches concentrate on single-view networks, that can only characterize one simple type of proximity relationships among objects. However, most of the real-world complex systems possess multiple types of relationships among entities. In this paper, a novel approach of graph embedding for multi-view networks is proposed, named Multi-view Graph Attention Networks (MGAT). We explore an attention-based architecture for learning node representations from each single view, the network parameters of which are constrained by a novel regularization term. In order to collaboratively integrate multiple types of relationships in different views, a view-focused attention method is explored to aggregate the view-wise node representations. We evaluate the proposed algorithm on several real-world datasets, and it demonstrates that the proposed approach outperforms existing state-of-the-art baselines.

Preserving differential privacy in deep neural networks with relevance-based adaptive noise imposition.

In recent years, deep learning achieves remarkable results in the field of artificial intelligence. However, the training process of deep neural networks may cause the leakage of individual privacy. Given the model and some background information of the target individual, the adversary can maliciously infer the sensitive feature of the target individual. Therefore, it is imperative to preserve the sensitive information in the training data. Differential privacy is a state-of-the-art paradigm for providing the privacy guarantee of datasets, which protects the private and sensitive information from the attack of adversaries significantly. However, the existing privacy-preserving models based on differential privacy are less than satisfactory since traditional approaches always inject the same amount of noise into parameters to preserve the sensitive information, which may impact the trade-off between the model utility and the privacy guarantee of training data. In this paper, we present a general differentially private deep neural networks learning framework based on relevance analysis, which aims to bridge the gap between private and non-private models while providing an effective privacy guarantee of sensitive information. The proposed model perturbs gradients according to the relevance between neurons in different layers and the model output. Specifically, during the process of backward propagation, more noise is added to gradients of neurons that have less relevance to the model output, and vice-versa. Experiments on five real datasets demonstrate that our mechanism not only bridges the gap between private and non-private models, but also prevents the disclosure of sensitive information effectively.

Local distinguishability aggrandizing network for human anomaly detection.

With the growing demand for an intelligent system to prevent abnormal events, many methods have been proposed to detect and locate anomalous behaviors in surveillance videos. However, most of these methods contain two shortcomings mainly: distraction of the network and insufficient discriminating ability. In this paper, we propose a local distinguishability aggrandizing network (LDA-Net) in a supervised manner, consisting of a human detection module and an anomaly detection module. In the human detection module, we obtain segmented patches of specific human subjects and take them as the input of the latter module to focus the network on learning motion characteristics of each person. In addition, considering that the auxiliary information, such as the specific type of an action, can aggrandize the whole network to extract distinguishable detail features of normal and abnormal behaviors, the proposed anomaly detection module comprises a primary binary classification sub-branch and an auxiliary distinguishability aggrandizing sub-branch, through which we can jointly detect anomalies and recognize actions. To further reduce the misclassification of the extremely imbalanced datasets, we design a novel inhibition loss function and embed it into the auxiliary sub-branch of the anomaly detection module. Experiments on several public benchmark datasets for frame-level and pixel-level anomaly detection show that the proposed supervised LDA-Net achieves state-of-the-art results on UCSD Ped2 and Subway Exit datasets.