Pairwise Embedding Research Articles

Global pixel-value-ordering framework with dynamic sequence partition for reversible data hiding

Recently, pixel-value-ordering (PVO) based reversible data hiding (RDH) methods have become hotspots in spatial domain RDH research. In PVO-based methods, the pixel value correlation within a local region is exploited for data embedding. To embed secret data, the cover images are partitioned into non-overlapped rectangle blocks to be sorted locally, and a lower embedding distortion is thus obtained. However, their embedding capacities are often limited by the fixed local blocks from which pixels are sorted. To maintain the advantages and overcome disadvantages of the PVO-based methods, a new global pixel-value-ordering (GPVO) framework is proposed in this paper. By applying our GPVO framework, a PVO-based method can utilize pixels from any position in the cover image rather than from local pixel blocks. Then, through the dynamic sequence partition realized by the GPVO framework for the first time, the embedding capacity of PVO-based methods can be greatly improved while reducing embedding distortion further reduced. Finally, we propose a two-stage pairwise embedding scheme to be applied to the sequences, aiming to achieve an advanced embedding performance. Experimental results illustrate that the proposed GPVO works better than other related state-of-the-art RDH methods.

Causal lifting and link prediction

Existing causal models for link prediction assume an underlying set of inherent node factors—an innate characteristic defined at the node’s birth—that governs the causal evolution of links in the graph. In some causal tasks, however, link formation is path-dependent : the outcome of link interventions depends on existing links. Unfortunately, these existing causal methods are not designed for path-dependent link formation, as the cascading functional dependencies between links (arising from path dependence ) are either unidentifiable or require an impractical number of control variables. To overcome this, we develop the first causal model capable of dealing with path dependencies in link prediction. In this work, we introduce the concept of causal lifting, an invariance in causal models of independent interest that, on graphs, allows the identification of causal link prediction queries using limited interventional data. Further, we show how structural pairwise embeddings exhibit lower bias and correctly represent the task’s causal structure, as opposed to existing node embeddings, e.g. graph neural network node embeddings and matrix factorization. Finally, we validate our theoretical findings on three scenarios for causal link prediction tasks: knowledge base completion, covariance matrix estimation and consumer-product recommendations.

Adversarial Neural Collaborative Filtering with Embedding Dimension Correlations

ABSTRACT Recently, convolutional neural networks (CNNs) have achieved excellent performance for the recommendation system by extracting deep features and building collaborative filtering models. However, CNNs have been verified susceptible to adversarial examples. This is because adversarial samples are subtle non-random disturbances, which indicates that machine learning models produce incorrect outputs. Therefore, we propose a novel model of Adversarial Neural Collaborative Filtering with Embedding Dimension Correlations, named ANCF in short, to address the adversarial problem of CNN-based recommendation system. In particular, the proposed ANCF model adopts the matrix factorization to train the adversarial personalized ranking in the prediction layer. This is because matrix factorization supposes that the linear interaction of the latent factors, which are captured between the user and the item, can describe the observable feedback, thus the proposed ANCF model can learn more complicated representation of their latent factors to improve the performance of recommendation. In addition, the ANCF model utilizes the outer product instead of the inner product or concatenation to learn explicitly pairwise embedding dimensional correlations and obtain the interaction map from which CNNs can utilize its strengths to learn high-order correlations. As a result, the proposed ANCF model can improve the robustness performance by the adversarial personalized ranking, and obtain more information by encoding correlations between different embedding layers. Experimental results carried out on three public datasets demonstrate that the ANCF model outperforms other existing recommendation models.

Local-Moment-Driven Robust Reversible Data Hiding

In this paper, a local-moment-driven robust reversible data hiding (LM-RRDH) scheme is proposed, which can provide security to hidden messages against unintentional modifications. The proposed LM-RRDH decomposes an image into LSB and MSB planes and then embeds the secret information into the MSB image so that intrusion by unintentional modifications can be avoided. In addition, the proposed scheme utilizes the prevalent correlation among the pixels on the MSB plane for optimal embedding. In the proposed scheme, a cover image is partitioned into sub-blocks at first, and pixel groups in the sub-block are formed according to local moment and moment-of-moment so that similar-intensity pixels can be grouped into the same group. Next, the secret data is embedded into the pixels of each group by selecting a pairwise embedding strategy adaptively which is based on the number of pixels in each group. As a result, the proposed LM-RRDH can limit the distortion while providing a decent embedding capacity. Further, a protection against non-malicious attacks such as Joint Photographic Experts Group (JPEG) compression is also provided. The experimental results show that the proposed scheme provides a superior quality to the previous works while providing a comparable embedding capacity.

Flexible patch moving modes for pixel-value-ordering based reversible data hiding methods

Pixel-value-ordering (PVO) is one of the most popular frameworks in the research of reversible data hiding (RDH) in recent years. In most PVO-based methods, the cover image is divided into non-overlapped blocks to embed secret data into the maximum and minimum pixels of each block. In these methods, the image division processing is just like moving a patch with constant step lengths in both horizontal and vertical directions. As a result, the block amount, as well as the embedding capacity (EC), is limited by the image size and the patch size. In this paper, we propose one-dimensional and two-dimensional flexible patch moving (FPM) modes to move the patch with adaptive step lengths. After enlarging the EC by employing FPM, an average-difference-based complexity computation is also proposed to cooperate with FPM to further improve the embedding performance. Finally, multiple pairwise embedding schemes are adaptively applied on blocks with different complexities to enhance the embedding performance once again. Experimental results illustrate that the proposed FPM modes work well on many PVO-based methods. Moreover, the proposed method achieves an obvious improvement in fidelity when compared with some state-of-the-art RDH schemes.

Scalable semantic-enhanced supervised hashing for cross-modal retrieval

Cross-modal retrieval is a classic scenario that can describe the same semantics from multiple angles and provide us with more abundant and diversified information ; thus, it has abundant applications in information retrieval, data mining, and machine learning. Generally, supervised information can enhance the ability to learn hash features, and obtain better retrieval accuracy than unsupervised methods. However, how to effectively solve the discrete binary constraint problem, and how to retain more semantic information in the process of hash learning, which constitute important parts of cross-modal retrieval, still have some challenges. To mitigate these problems, in this work, we propose a novel cross-modal hashing approach, called, scalable pairwise embedding constraint hashing (SPECH). The SPECH approach employs the loss of likelihood similarity technique using pairwise sample data to measure the semantic similarity of heterogeneous modal samples. In light of this, the SPECH approach can maximize the utilization of available data from heterogeneous modal samples for training models. In addition, we further enhance the discriminative capability of hash codes and reduce the intraclass differences in heterogeneous modal samples, by embedding the more discriminative semantic label attribute space regression into the to-be-learned process of hash codes. Comprehensive experiments on three benchmark datasets show that the proposed SPECH approach achieves significantly better retrieval accuracy and outperforms several state-of-the-art approaches on cross-modal retrieval applications.

Robust RDH Technique using Sorting and IPVO-based Pairwise PEE for Secure Communication

Nowadays, Reversible Data Hiding (RDH) is used extensively in information sensitive communication domains to protect the integrity of hidden data and the cover medium. However, most of the recently proposed RDH methods lack robustness. Robust RDH methods are required to protect the hidden data from security attacks at the time of communication between the sender and receiver. In this paper, we propose a Robust RDH scheme using IPVO based pairwise embedding. The proposed scheme is designed to prevent unintentional modifications caused to the secret data by JPEG compression. The cover image is decomposed into two planes namely HSB plane and LSB plane. As JPEG compression most likely modifies the LSBs of the cover image during compression, it is best not to hide the secret data into LSB planes. So, the proposed method utilizes a pairwise embedding to embed secret data into HSB plane of the cover image. High fidelity improved pixel value ordering (IPVO) based pairwise embedding ensures that the embedding performance of the proposed method is improved.

Robust RDH Technique Using Sorting and IPVO-Based Pairwise PEE for Secure Communication

Nowadays, Reversible Data Hiding (RDH) is used extensively in information sensitive communication domains to protect the integrity of hidden data and the cover medium. However, most of the recently proposed RDH methods lack robustness. Robust RDH methods are required to protect the hidden data from security attacks at the time of communication between the sender and receiver. In this paper, we propose a Robust RDH scheme using IPVO based pairwise embedding. The proposed scheme is designed to prevent unintentional modifications caused to the secret data by JPEG compression. The cover image is decomposed into two planes namely HSB plane and LSB plane. As JPEG compression most likely modifies the LSBs of the cover image during compression, it is best not to hide the secret data into LSB planes. So, the proposed method utilizes a pairwise embedding to embed secret data into HSB plane of the cover image. High fidelity improved pixel value ordering (IPVO) based pairwise embedding ensures that the embedding performance of the proposed method is improved.

The Evaluation of Musical Influence and Evolution

Music, which has evolved for thousands of years, plays an important role in social culture. To understand musical evolution, we need to measure the influence of previous music on new music and examine the revolutionary shift over the past years. We prepare the sets of artists, songs and genres as 3 basic types of music carriers. Together with music features and time, we have several 3-dimentional attribute matrices describing the musical properties during different time spans for artists and songs. Artists, songs and genres also have pairwise embeddings and they have internal one-to-one influencer-to-follower relationships.

Pairwise Link Prediction Model for Out of Vocabulary Knowledge Base Entities

Real-world knowledge bases such as DBPedia, Yago, and Freebase contain sparse linkage connectivity, which poses a severe challenge to link prediction between entities. To cope with such data scarcity issues, recent models have focused on learning interactions between entity pairs by means of relations that exist between them. However promising, some relations are associated with very few tail entities or head entities, resulting in poor estimation of the relation interaction between entities. In this article, we break the sole dependency of modeling relation interactions between entity pairs by associating a triple with pairwise embeddings, i.e., distributed vector representations for pairs of word-based entities and relation of a triple. We capture the interactions that exist between pairwise embeddings by means of a Pairwise Factorization Model that employs a factorization machine with relation attention. This approach allows parameters for related interactions to be estimated efficiently, ensuring that the pairwise embeddings are discriminative, providing strong supervisory signals for the decoding task of link prediction. The Pairwise Factorization Model we propose exploits a neural bag-of-words model as the encoder, which effectively encodes word-based entities into distributed vector representations for the decoder. The proposed model is simple and enjoys efficiency and capability, showing superior link prediction performance over state-of-the-art complex models on benchmark datasets DBPedia50K and FB15K-237.

Enhanced pairwise IPVO-based reversible data hiding scheme using rhombus context

Pixel-Value-Ordering (PVO) based reversible data hiding schemes provide high-fidelity stego-images with moderate embedding capacity. In this paper, a novel reversible data hiding scheme based on enhanced pairwise pixel value ordering is proposed to increase embedding capacity. The proposed scheme divides a host image into non-overlapping blocks of three pixels in zig-zag order and categorizes the blocks into two categories namely absolute smooth and probably complex based on the rhombus mean for each pixel of the block. In the case of absolute smooth, the pairwise embedding is done in two-pass using two different predictors. In the pass-1, enhanced pairwise improved PVO method is used to embed the secret data in the block. The proposed method decreases the minimum valued pixels and increases the maximum valued pixels for embedding the secret data. In the pass-2, secret data is hidden using a novel recovery based pairwise embedding strategy, where the value of the first pixel according to the sorted mean sequence will be either increased or remains unchanged while the value of the last pixel will be either decreased or remains unchanged. Additionally, the medium pixel is also used to embed the secret data by modifying its value in any direction using prediction error expansion. Therefore, the pass-2 embedding is done in such a way that it becomes the complement of pass-1 embedding so that stego-image quality can be maintained with additional embedded data. In case of probable complex blocks, the secret data is embedded in pixel wise manner to local complexity of each pixel. The use of rhombus mean in block categorization allows embedding in each and every pixel of the host image while ensuring its reversibility. Thus, the proposed scheme significantly increases the embedding capacity. The experimental results show that the proposed scheme has almost doubled on the embedding capacity than the previous PVO-based reversible data hiding schemes. Further, it has higher PSNR at high embedding capacity.

Improved reversible data hiding based on PVO and adaptive pairwise embedding

Pixel-value-ordering (PVO) is an efficient technique of reversible data hiding (RDH). By PVO, the cover image is first divided into non-overlapping blocks with equal size. Then, the pixel values in each block are sorted in ascending order. Next, take the second largest/samllest pixel value as a prediction of the largest/samllest pixel value to derive two prediction errors. Finally, the data embedding is constructed by modifying the generated prediction errors of each block. After data embedding, the PVO of each block is unchanged, which guarantees the reversibility. Our key observation is that, in each block, the modification for the two prediction errors is independent without exploiting the correlation between them, although they are closely correlated with each other. In light of this, an improved PVO-based RDH method is proposed in this work. The two prediction errors of each block are considered as a pair, and the pairs are modified for data embedding based on adaptive two-dimensional histogram modification. The proposed method is experimentally verified better than the original PVO-based method and some of its improvements.

Reversible data hiding based on pairwise embedding and optimal expansion path

Abstract As an efficient technique for high-dimensional reversible data hiding (RDH), pairwise prediction-error expansion (pairwise PEE) has achieved better performance comparing with the conventional PEE. With pairwise PEE, the correlations among prediction-errors are well utilized by modifying the generated two-dimensional prediction-error histogram (2D-PEH). However, its performance can be further improved since the histogram modification manner (i.e., the employed modification mapping) of pairwise PEE is fixed and independent of image content. To better utilize image redundancy, instead of embedding data based on an empirically designed modification mapping, a content dependent pairwise embedding scheme is proposed in this paper. Based on a specific division of 2D-PEH, the expansion bins selection is formulated as an optimal path determination problem, and the histogram modification mapping is adaptively determined by taking the optimal expansion bins. To reduce the computation cost, a dynamic programming algorithm is proposed to solve the optimization problem with low computational complexity. Moreover, by combining the proposed optimal expansion path with the existing one-dimensional adaptive embedding mechanism, the embedding performance can be further enhanced. The proposed method performs well and its superiority is experimentally verified comparing with pairwise PEE and some other state-of-the-art methods.

Adaptive image data hiding in edges using patched reference table and pair-wise embedding technique

Reference table (RT) based embedding method embeds secret digits into pixel pairs under the guidance of a reference table. Most of the existing RT-based methods either require elaborate conversions among different bases, or have limited embedding capacity. In this paper, we use a patched reference table (PRT) as a guide and propose a PRT method to provide a better image quality and extendable embedding capacity. We also exploit the concept of pixel value differencing (PVD) and propose another method PRT–PVD. In the traditional PVD-based methods, the shape of the difference histograms of the stego images is significantly altered and, thus, vulnerable to some steganalyzers. PRT–PVD adopts the PRT method and uses a specially designed embedding sequence to preserve the difference histogram shape. Experimental results reveal that the proposed PRT and PRT–PVD methods not only have better embedding efficiency over the existing methods, but also are robust to detection by modern steganalysis tools.