Language Bottleneck Models (LBMs) are proposed to achieve interpretable image recognition by classifying images based on textual concept bottlenecks. However, current LBMs simply list all concepts together as the bottleneck layer, leading to the spurious cue inference problem and cannot generalized to unseen classes. To address these limitations, we propose the Attribute-formed Language Bottleneck Model (ALBM). ALBM organizes concepts in the attribute-formed class-specific space, where concepts are descriptions of specific attributes for specific classes. In this way, ALBM can avoid the spurious cue inference problem by classifying solely based on the essential concepts of each class. In addition, the cross-class unified attribute set also ensures that the concept spaces of different classes have strong correlations, as a result, the learned concept classifier can be easily generalized to unseen classes. Moreover, to further improve interpretability, we propose Visual Attribute Prompt Learning (VAPL) to extract visual features on fine-grained attributes. Furthermore, to avoid labor-intensive concept annotation, we propose the Description, Summary, and Supplement (DSS) strategy to automatically generate high-quality concept sets with a complete and precise attribute. Extensive experiments on 9 widely used few-shot benchmarks demonstrate the interpretability, transferability, and performance of our approach. The code and collected concept sets are available at https://github.com/tiggers23/ALBM.

Damage analysis in social media platforms such as Twitter is a comprehensive problem which involves different subtasks for mining damage-related information from tweets ( e.g., informativeness, humanitarian categories and severity assessment). The comprehensive information obtained by damage analysis enables to identify breaking events around the world in real-time and hence provides aids in emergency responses. Recently, with the rapid development of web technologies, multimodal damage analysis has received increasing attentions due to users' preference of posting multimodal information in social media. Multimodal damage analysis leverages the associated image modality to improve the identification of damage-related information in social media. However, existing works on multimodal damage analysis address each damage-related subtask individually and do not consider their joint training mechanism. In this work, we propose the Bidirectional Multi-task Cascaded multimodal Fusion (BiMCF) approach towards joint multimodal damage analysis. To this end, we introduce the cascaded multimodal fusion framework to separately integrate effective visual and text information for each task, considering that different tasks attend to different information. To exploit the interactions across tasks, bidirectional propagation of the attended image-text interactive information is implemented between tasks, which can lead to enhanced multimodal fusion. Comprehensive experiments are conducted to validate the effectiveness of the proposed approach. Code is available at https://github.com/tiggers23/BiMCF.

Partial-label learning (PLL) allows each training example to be equipped with a set of candidate labels. Existing deep PLL research focuses on a \emph{learning-centric} perspective to design various training strategies for label disambiguation i.e., identifying the concealed true label from the candidate label set, for model training. However, when the size of the candidate label set becomes excessively large, these learning-centric strategies would be unable to find the true label for model training, thereby causing performance degradation. This motivates us to think from a \emph{data-centric} perspective and pioneer a new PLL-related task called candidate label set pruning (CLSP) that aims to filter out certain potential false candidate labels in a training-free manner. To this end, we propose the first CLSP method based on the inconsistency between the representation space and the candidate label space. Specifically, for each candidate label of a training instance, if it is not a candidate label of the instance's nearest neighbors in the representation space, then it has a high probability of being a false label. Based on this intuition, we employ a per-example pruning scheme that filters out a specific proportion of high-probability false candidate labels. Theoretically, we prove an upper bound of the pruning error rate and analyze how the quality of representations affects our proposed method. Empirically, extensive experiments on both benchmark-simulated and real-world PLL datasets validate the great value of CLSP to significantly improve many state-of-the-art deep PLL methods.

Large pre-trained vision-language models like CLIP have shown amazing zero-shot recognition performance. To adapt pre-trained vision-language models to downstream tasks, recent studies have focused on the learnable context + class name paradigm, which learns continuous prompt contexts on downstream datasets. In practice, the learned prompt context tends to overfit the base categories and cannot generalize well to novel categories out of the training data. Recent works have also noticed this problem and have proposed several improvements. In this work, we draw a new insight based on empirical analysis, that is, uninformative class names lead to degraded base-to-novel generalization performance in prompt learning, which is usually overlooked by existing works. Under this motivation, we advocate to improve the base-to-novel generalization performance of prompt learning by enhancing the semantic richness of class names. We coin our approach as the Information Disengagement based Associative Prompt Learning (IDAPL) mechanism which considers the associative, meanwhile, decoupled learning of prompt context and class name embedding. IDAPL can effectively alleviate the phenomenon of learnable context overfitting to base classes, meanwhile, learning more informative semantic representation of base classes by fine-tuning the class name embedding, leading to improved performance on both base and novel classes. Experimental results on eleven widely used few-shot learning benchmarks clearly validate the effectiveness of our proposed approach. Code is available at https://github.com/tiggers23/IDAPL

In many real-world machine learning classification applications, the model performance based on deep neural networks (DNNs) oftentimes suffers from label noise. Various methods have been proposed in the literature to address this issue, primarily by focusing on designing noise-tolerant loss functions, cleaning label noise, and correcting the objective loss. However, the noise-tolerant loss functions face challenges when the noise level increases. This article aims to reveal a convergence path of a trained model in the presence of label noise, and here, the convergence path depicts the evolution of a trained model over epochs. We first propose a theorem to demonstrate that any surrogate loss function can be used to learn DNNs from noisy labels. Next, theories on the general convergence path for the deep models under label noise are presented and verified through a series of experiments. In addition, we design an algorithm based on the proposed theorems that make efficient corrections on the noisy labels and achieve strong robustness in the DNN models. We designed several experiments using benchmark datasets to assess noise tolerance and verify the theorems presented in this article. The comprehensive experimental results firmly confirm our theoretical results and also clearly validate the effectiveness of our method under various levels of label noise.

In partial-label learning (PLL), each training example has a set of candidate labels, among which only one is the true label. Most existing PLL studies focus on the instance-independent (II) case, where the generation of candidate labels is only dependent on the true label. However, this II-PLL paradigm could be unrealistic, since candidate labels are usually generated according to the specific features of the instance. Therefore, instance-dependent PLL (ID-PLL) has attracted increasing attention recently. Unfortunately, existing ID-PLL studies lack an insightful perception of the intrinsic challenge in ID-PLL. In this paper, we start with an empirical study of the dynamics of label disambiguation in both II-PLL and ID-PLL. We found that the performance degradation of ID-PLL stems from the inaccurate supervision caused by massive under-disambiguated (UD) examples that do not achieve complete disambiguation. To solve this problem, we propose a novel two-stage PLL framework including selective disambiguation and candidate-aware thresholding. Specifically, we first choose a part of well-disambiguated (WD) examples based on the magnitude of normalized entropy (NE) and integrate harmless complementary supervision from the remaining ones to train two networks. Next, the remaining examples whose NE is lower than the specific class-wise WD-NE threshold are selected as additional WD ones. Meanwhile, the remaining UD examples, whose NE is lower than the self-adaptive UD-NE threshold and whose predictions from two networks are agreed, are also regarded as WD ones for model training. Extensive experiments demonstrate that our proposed method outperforms state-of-the-art PLL methods.

Partial-label learning (PLL) relies on a key assumption that the true label of each training example must be in the candidate label set. This restrictive assumption may be violated in complex real-world scenarios, and thus the true label of some collected examples could be unexpectedly outside the assigned candidate label set. In this paper, we term the examples whose true label is outside the candidate label set OOC (Out-Of-Candidate) examples, and pioneer a new PLL study to learn with OOC examples. We consider two types of OOC examples in reality, i.e., the closed-set/open-set OOC examples whose true label is inside/outside the known label space. To solve this new PLL problem, we first calculate the wooden cross-entropy loss from candidate and non-candidate labels respectively, and dynamically differentiate the two types of OOC examples based on specially designed criteria. Then, for closed-set OOC examples, we conduct reversed label disambiguation in the non-candidate label set; for open-set OOC examples, we leverage them for training by utilizing an effective regularization strategy that dynamically assigns random candidate labels from the candidate label set. In this way, the two types of OOC examples can be differentiated and further leveraged for model training. Extensive experiments demonstrate that our proposed method outperforms state-of-the-art PLL methods.

Abstract Affine equivalence of Boolean functions has various applications in computer science and modern cryptography, such as circuit design and S-boxes. Existing methods for detecting affine equivalence of Boolean functions work in some cases but not when the truth table of a Boolean function is sparse. To improve previous methods and overcome this limitation, we propose a method by transforming the Boolean function to a function with the property that its function values at the orthonormal basis are all equal to 1 or 0, which narrows down the search space of affine transformations. Our first algorithm has the advantage of getting a smaller search space than previous methods and is especially useful for sparse functions. Specifically, when the Boolean functions are sparse, the search space can be reduced exponentially in average and experiments show the efficiency of our first algorithm. We then present another algorithm to transform one circuit into its equivalent affine circuit by synthesizing a reversible circuit and inserting it in front of the original circuit. To our knowledge, this is the first work to automatically synthesize an affine equivalent circuit for any given circuit and the first to do this by combining reversible circuit and non-reversible circuit.

Deep models have achieved state-of-the-art performance on a broad range of visual recognition tasks. Nevertheless, the generalization ability of deep models is seriously affected by noisy labels. Though deep learning packages have different losses, this is not transparent for users to choose consistent losses. This paper addresses the problem of how to use abundant loss functions designed for the traditional classification problem in the presence of label noise. We present a dynamic label learning (DLL) algorithm for noisy label learning and then prove that any surrogate loss function can be used for classification with noisy labels by using our proposed algorithm, with a consistency guarantee that the label noise does not ultimately hinder the search for the optimal classifier of the noise-free sample. In addition, we provide a depth theoretical analysis of our algorithm to verify the justifies' correctness and explain the powerful robustness. Finally, experimental results on synthetic and real datasets confirm the efficiency of our algorithm and the correctness of our justifies and show that our proposed algorithm significantly outperforms or is comparable to current state-of-the-art counterparts.

We consider the problem of checking the generalized affine equivalence of two given Boolean functions. This problem arises in various computer-aided design (CAD) and cryptographic applications, such as circuit synthesis and Reed–Muller codes. Based on the theory of affine group acting on the Boolean functions, we define the coefficient spaces and transition relations, and transform the checking problem into reachability analysis of finite state machines. Two methods are proposed to check the affine equivalence of Boolean functions using binary decision diagrams (BDDs) and property directed reachability (PDR), respectively. Both methods can check affine equivalence of bent and semi-bent functions, which state-of-the-art methods can hardly handle. Furthermore, existing methods only consider the case of affine equivalence, while our methods can handle the generalized affine equivalence of subspaces of Boolean functions. In the application of circuit synthesis, our methods can significantly reduce the size of the library compared to Boolean matching. To classify Boolean functions up to the generalized affine equivalence, we propose a method to obtain a complete classification based on BDDs. In the experiments, we have successfully applied our methods to some examples that can hardly be solved by using the previous methods, thus, validating the effectiveness of our methods.

Unsupervised domain adaptation has limitations when encountering label discrepancy between the source and target domains. While open-set domain adaptation approaches can address situations when the target domain has additional categories, these methods can only detect them but not further classify them. In this paper, we focus on a more challenging setting dubbed Domain Adaptive Zero-Shot Learning (DAZSL), which uses semantic embeddings of class tags as the bridge between seen and unseen classes to learn the classifier for recognizing all categories in the target domain when only the supervision of seen categories in the source domain is available. The main challenge of DAZSL is to perform knowledge transfer across categories and domain styles simultaneously. To this end, we propose a novel end-to-end learning mechanism dubbed Three-way Semantic Consistent Embedding (TSCE) to embed the source domain, target domain, and semantic space into a shared space. Specifically, TSCE learns domain-irrelevant categorical prototypes from the semantic embedding of class tags and uses them as the pivots of the shared space. The source domain features are aligned with the prototypes via their supervised information. On the other hand, the mutual information maximization mechanism is introduced to push the target domain features and prototypes towards each other. By this way, our approach can align domain differences between source and target images, as well as promote knowledge transfer towards unseen classes. Moreover, as there is no supervision in the target domain, the shared space may suffer from the catastrophic forgetting problem. Hence, we further propose a ranking-based embedding alignment mechanism to maintain the consistency between the semantic space and the shared space. Experimental results on both I2AwA and I2WebV clearly validate the effectiveness of our method. Code is available at https://github.com/tiggers23/TSCE-Domain-Adaptive-Zero-Shot-Learning.

Existing sentiment analysis models have achieved great advances with the help of sufficient sentiment annotations.Unfortunately, many languages do not have sufficient sentiment corpus.To this end, recent studies have proposed cross-lingual sentiment analysis to transfer sentiment analysis models from resource-rich languages to low-resource languages.However, these studies either rely on external cross-lingual supervision (e.g., parallel corpora and translation model), or are limited by the cross-lingual gaps.In this work, based on the intuitive assumption that the relationships between emojis and sentiments are consistent across different languages, we investigate transferring sentiment knowledge across languages with the help of emojis.To this end, we propose a novel cross-lingual sentiment analysis approach dubbed Curriculum Knowledge Distiller (CKD).The core idea of CKD is to use emojis to bridge the source and target languages.Note that, compared with texts, emojis are more transferable, but cannot reveal the precise sentiment.Thus, we distill multiple Intermediate Sentiment Classifiers (ISC) on source language corpus with emojis to get ISCs with different attention weights of texts.To transfer them into the target language, we distill ISCs into the Target Language Sentiment Classifier (TSC) following the curriculum learning mechanism.In this way, TSC can learn delicate sentiment knowledge, meanwhile, avoid being affected by cross-lingual gaps.Experimental results on five cross-lingual benchmarks clearly verify the effectiveness of our approach.

Partial-label learning (PLL) solves the problem where each training instance is assigned a candidate label set, among which only one is the ground-truth label. The core of PLL is to learn efficient feature representations to facilitate label disambiguation. However, existing PLL methods only learn plain representations by coarse supervision, which is incapable of capturing sufficiently distinguishable representations, especially when confronted with the knotty label ambiguity, i.e., certain candidate labels share similar visual patterns. In this paper, we propose a novel framework partial label learning with semantic label representations dubbed ParSE, which consists of two synergistic processes, including visual-semantic representation learning and powerful label disambiguation. In the former process, we propose a novel weighted calibration rank loss that has two implications. First, it implies a progressive calibration strategy that utilizes the disambiguated label confidence to weight the similarity between each image feature embedding and its corresponding semantic label representations of all candidates. Second, it also considers the ranking relationship between candidate and non-candidate ones. Based on learned visual-semantic representations, subsequent label disambiguation is desirably endowed with more powerful abilities. Experiments on benchmarks show that ParSE outperforms state-of-the-art counterparts.

Massive physical devices are deployed in the Industrial Internet of Things (IoT) to collect ambiance data while heavy storage and communication cost are imposed on these IoT devices. To overcome this constraint, cloud-assisted technologies are introduced to store and manage the collected data. In order to protect data quality and security, encryption is required before uploading data to remote clouds. Consequently, a search function is added to cloud services to find the specific data. However, traditional data searching schemes are constructed in user-oriented systems, where the search function is mainly involved with the relationship between data and users rather than data and devices. As a result, traditional search schemes are not suitable to find special IoT devices. On the other hand, the status of these devices is described by many attributes, e.g., temperature, clean water storage, and machine speed in an early warning system for industrial sewage disposal equipment. Hence, multi-keyword conjunctive queries for partial attributes should be introduced so as to find the target device more accurately and more efficiently. To address these challenges, we propose a new universal device-oriented keyword searchable encryption (Do-KSE) scheme for cloud-assisted IoT in this paper. Furthermore, the functions of a single and conjunctive keyword search are maintained to handle the device search requirement of partial attributes. We conduct extensive experiments to evaluate the proposed scheme. Experimental results show that our scheme has excellent performance because of the lightweight index and query trapdoor.

Recommendation efficiency and data sparsity problems have been regarded as two main challenges of real-world recommendation systems. Most existing works focus on improving recommendation accuracy instead of efficiency. In this paper, we propose a Deep Pairwise Hashing (DPH) to map users and items to binary vectors in the Hamming space, where a user's preference for an item can be efficiently calculated by the Hamming distance, which significantly improves the efficiency of online recommendation. To alleviate data sparsity and cold-start problems, the item content information exploited and integrated to learn effective representations of items. Specifically, we first pre-train robust item representation from item content data by a robust Denoising Auto-encoder instead of other deterministic deep learning frameworks. Then we fine-tune the entire recommender framework by adding a pairwise loss function with discrete constraints, which is more consistent with the ultimate goal of producing a ranked list of items. Finally, we adopt the alternating optimization method to optimize the proposed model with discrete constraints. Extensive experiments conducted on three different datasets show that DPH can significantly advance the state-of-the-art frameworks regarding data sparsity and cold-start item recommendation.

Deep learning has achieved tremendous success in image classification. However, the corresponding performance leap relies heavily on large-scale accurate annotations, which are usually hard to collect in reality. It is essential to explore methods that can train deep models effectively under label noise. To address the problem, we propose to train deep models with robust binary loss functions. To be specific, we tackle the K-class classification task by using K binary classifiers. We can immediately use multi-category large margin classification approaches, e.g., Pairwise-Comparison (PC) or One-Versus-All (OVA), to jointly train the binary classifiers for multi-category classification. Our method can be robust to label noise if symmetric functions, e.g., the sigmoid loss or the ramp loss, are employed as the binary loss function in the framework of risk minimization. The learning theory reveals that our method can be inherently tolerant to label noise in multi-category classification tasks. Extensive experiments over different datasets with different types of label noise are conducted. The experimental results clearly confirm the effectiveness of our method.

Smart city, as a new mode, is introduced to improve the level of city management for modern cities. In smart cities, a kernel field is health management for urban residents. Hospital management, as one of the most important components in health management, is concerned. To provide high-quality medical service for sick residents, accurate patient health data analysis is needed. Thus, data collection in patient health monitoring is necessary. To achieve this, massive Internet-of-Things devices are distributed; in general, they are resource-constrained devices. From this, lightweight index generation is needed. Furthermore, with the development of professional technologies in medical science, the hospital manager has to employ many different types of professional doctors. They need the shared patient health data to do a precise diagnosis and present an efficient therapeutic schedule for each patient. However, many secret details are recorded in the patient health data. Thus, data privacy of the shared patient health data should be maintained. In this article, we propose a new traceable patient health data search system for hospital management in smart cities. In this system, the system manager shares the encrypted patient health data to different doctors at the grain of hospital bed. Each doctor accurately finds a patient with a special feature from the patient health monitoring data. To prevent patient health data leakage, the functions of illegal search query blocking and inside malicious user tracing are designed. The performance analysis shows that our system is practical for lightweight data collecting devices.

Deep Neural Networks have provided excellent performance in a variety of applications. However, their superior performance comes with a huge expense of collecting a correctly-annotated large-scale training set, and a little impracticality of preparing such a training set in many applications. In this work, we study another natural type of weak supervision, complementary-label learning, to address this problem. Complementary-label learning refers to train the Deep Neural Networks by the usage of only complementary labels, and a complementary label indicates one of the classes that the sample does not belong to. This paper first presents a general risk formulation for complementary label learning through an adoption of arbitrary losses designed for ordinary-label learning. We then theoretically analyze that our method is applicable for any loss functions to learn deep neural networks with complementary labels in the framework of risk minimization. Experimental results on different benchmark datasets demonstrate that our approach outperforms current state-of-the-art methods.

Parallel adders are extensively used in high-performance computer design and hardware acceleration for large-scale data processing. In the adder design theory, a key property of the group propagated carry and the group generated carry is based on the two recurrence equations. The property is fundamental to many parallel prefix adders. However, there is no proof of the property in the literature. This article presents a rigorous and complete proof for it. The proof can leverage a solid ground for a formal verification methodology for parallel adder-based chip design.

Semantic segmentation, which aims to acquire pixel-level understanding about images, is among the key components in computer vision. To train a good segmentation model for real-world images, it usually requires a huge amount of time and labor effort to obtain sufficient pixel-level annotations of real-world images beforehand. To get rid of such a nontrivial burden, one can use simulators to automatically generate synthetic images that inherently contain full pixel-level annotations and use them to train a segmentation model for the real-world images. However, training with synthetic images usually cannot lead to good performance due to the domain difference between the synthetic images (i.e., source domain) and the real-world images (i.e., target domain). To deal with this issue, a number of unsupervised domain adaptation (UDA) approaches have been proposed, where no labeled real-world images are available. Different from those methods, in this work, we conduct a pioneer attempt by using easy-to-collect image-level annotations for target images to improve the performance of cross-domain segmentation. Specifically, we leverage those image-level annotations to construct curriculums for the domain adaptation problem. The curriculums describe multi-level properties of the target domain, including label distributions over full images, local regions and single pixels. Since image annotations are “weak” labels compared to pixel annotations for segmentation, we coin this new problem as weakly-supervised cross-domain segmentation. Comprehensive experiments on the GTA5→ Cityscapes and SYNTHIA→ Cityscapes settings demonstrate the effectiveness of our method over the existing state-of-the-art baselines.

Current deep learning methods developed to address classification problems related to medical imaging for disease detection and diagnosis are primarily based on binary labels and also with limited focus on confidence calibration. Confidence estimates are closely related to classification accuracy. While existing neural networks have the capability of extending binary labels to multiclass labels, the confidence calibration procedure is generally overlooked. To address the issue, we propose a method called knowledge discriminator risk network (KDR) and a confidence calibration voting algorithm (KDR-CCV) that together enhance classification accuracy, with an emphasis on confidence calibration. Comparative studies on multiclass classification based on the Breast Imaging Reporting and Data Systems (BI-RADS) assessment categories with a dataset containing only binary labels of ultrasound images are conducted. Experimental results show KDR-CCV achieves the overall best classification performance in comparison to other methods that conform to the BI-RADS criterion in addition to the effective improvement on classification accuracy. The proposed method incorporates BI-RADS assessment and artificial intelligence from an application-based broad practice, and can be extended to other medical imaging problems.

Semantic segmentation aims to obtain a detailed understanding of images. Deep learning has achieved great advances in semantic segmentation over the past years. In practice, however, the classes do not always correspond to the ones in the training stage. Since it is impractical to collect sufficient labeled data for all classes, zero-shot semantic segmentation has received increasing attentions recently. Although semantic segmentation neural networks can transfer knowledge from seen classes to unseen classes by incorporating the class-level semantic information, it shows a strong bias towards seen classes. In this letter, we propose an easy-to-implement transductive approach to alleviate the prediction bias in zero-shot semantic segmentation. We assume that both source images with full pixel-level labels and unlabeled target images are available for training. The source images are used to build the relationship between visual images and class-level semantic embeddings. On the other hand, the target images are used to alleviate the bias towards seen classes. Comprehensive experiments over the PASCAL dataset clearly demonstrate the effectiveness of our approach.

Authentication protocol verification is a difficult problem. The problem of “state space explosion” has always been inevitable in the field of verification. Using inductive characteristics, we combine mathematical induction and model detection technology to solve the problem of “state space explosion” in verifying the OSK protocol and VOSK protocol of RFID system. In this paper, the security and privacy of protocols in RFID systems are studied and analysed to verify the effectiveness of the combination of mathematical induction and model detection. We design a (r,s,t)-security experiment on the basis of privacy experiments in the RFID system according to the IND-CPA security standard in cryptography, using mathematical induction to validate the OSK protocol and VOSK protocol. Finally, the following conclusions are presented. The OSK protocol cannot resist denial of service attacks or replay attacks. The VOSK protocol cannot resist denial of service attacks but can resist replay attacks. When there is no limit on communication, the OSK protocol and VOSK protocol possess (r,s,t)-privacy; that is to say they can resist denial of service attacks.

The efficiency of top-k recommendation is vital to large-scale recommender systems. Hashing is not only an efficient alternative but also complementary to distributed computing, and also a practical and effective option in a computing environment with limited resources. Hashing techniques improve the efficiency of online recommendation by representing users and items by binary codes. However, objective functions of existing methods are not consistent with ultimate goals of recommender systems, and are often optimized via discrete coordinate descent, easily getting stuck in a local optimum. To this end, we propose a Discrete Ranking-based Matrix Factorization (DRMF) algorithm based on each user's pairwise preferences, and formulate it into binary quadratic programming problems to learn binary codes. Due to non-convexity and binary constraints, we further propose self-paced learning for improving the optimization, to include pairwise preferences gradually from easy to complex. We finally evaluate the proposed algorithm on three public real-world datasets, and show that the proposed algorithm outperforms the state-of-the-art hashing-based recommendation algorithms, and even achieves comparable performance to matrix factorization methods.

Cold-start problem and recommendation efficiency have been regarded as two crucial challenges in the recommender system. In this paper, we propose a hashing based deep learning framework called Discrete Deep Learning (DDL), to map users and items to Hamming space, where a user»s preference for an item can be efficiently calculated by Hamming distance, and this computation scheme significantly improves the efficiency of online recommendation. Besides, DDL unifies the user-item interaction information and the item content information to overcome the issues of data sparsity and cold-start. To be more specific, to integrate content information into our DDL framework, a deep learning model, Deep Belief Network (DBN), is applied to extract effective item representation from the item content information. Besides, the framework imposes balance and irrelevant constraints on binary codes to derive compact but informative binary codes. Due to the discrete constraints in DDL, we propose an efficient alternating optimization method consisting of iteratively solving a series of mixed-integer programming subproblems. Extensive experiments have been conducted to evaluate the performance of our DDL framework on two different Amazon datasets, and the experimental results demonstrate the superiority of DDL over the state-of-the-art methods regarding online recommendation efficiency and cold-start recommendation accuracy.

Cloud storage has been used to reduce the cost and support convenient collaborations for industrial Internet of things (IIoT) data management. When data owners share IIoT data with authorized parties for data interaction, secure cloud data searching and file access control are fundamental security requirements. In this paper, first we discuss a new insider attack to the Cui's multi-key aggregate searchable encryption scheme, where the unauthorized inside users can guess the other users private keys. Then, we propose a novel file-centric multi-key aggregate keyword searchable encryption (Fc-MKA-KSE) system for the IIoT data in the file-centric framework. Specifically, we present two formal security models, namely, the security models of the indistinguishable selective-file chosen keyword attack and the indistinguishable selective-file keyword guessing attack, which can satisfy the security requirements. Our experimental results show that the proposed scheme achieves computational efficiency.

Personalized ranking is usually considered as an ultimate goal of recommendation systems, but it suffers from efficiency issues when making recommendations. To this end, we propose a learning-based hashing framework called Discrete Personalized Ranking (DPR), to map users and items to a Hamming space, where user-item affinity can be efficiently calculated via Hamming distance. Due to the existence of discrete constraints, it is possible to exploit a two-stage learning procedure for learning binary codes according to most existing methods. This two-stage procedure consists of relaxed optimization by discarding discrete constraints and subsequent binary quantization. However, such a procedure has been shown resulting in a large quantization loss, so that longer binary codes would be required. To this end, DPR directly tackles the discrete optimization problem of personalized ranking. And the balance and un-correlation constraints of binary codes are imposed to derive compact but informatics binary codes. Based on the evaluation on several datasets, the proposed framework shows consistent superiority to the competing baselines even though only using shorter binary code.

Recommendation is widely used to deal with information overloading by suggesting items based on historical information of users. One of the most popular recommendation techniques is matrix factorization (MF), in which the preferences of users are estimated by dot products of their real latent factors between users and items. Although MF can achieve high recommendation accuracy, it suffers from efficiency issues when making preferences ranking in real space. Hash retrieval technique can be applied to recommender systems to speed up preferences ranking. Due to the existence of discrete constraints in learning hash codes, it is possible to exploit a two-stage learning procedure according to most existing methods. This two-stage procedure consists of relaxed optimization by discarding discrete constraints and subsequent binary quantization. However, existing methods have not been able to well handle the change of dot product arising from quantization. To this end, we propose a dot-product based preference preserved hashing method, which quantizes both norm and cosine similarity in dot product respectively. We also design an algorithm to optimize the bit length for norm quantization. Based on the evaluation to several datasets, the proposed framework shows consistent superiority to the competing baselines even though only using shorter binary code.

Recommender systems have been widely used to deal with information overload, by suggesting relevant items that match users' personal interest. One of the most popular recommendation techniques is matrix factorization (MF). The inner products of learned latent factors between users and items can estimate users' preferences for items with high accuracy, but the preferences ranking is time consuming. Thus, hashing-based fast search technologies were exploited in recommender systems. However, most previous approaches consist of two stages: continuous latent factor learning and binary quantization, but they didn't well deal with the change of inner product arising from quantization. To this end, in this paper, we propose a constraint free preference preserving hashing method, which quantizes both norm and similarity in dot product. We also design an algorithm to optimize the bit length for norm quantization. The performance of our method is evaluated on three real world datasets. The results confirm that the proposed model can improve recommendation performance by 11%-15%, as compared with the state-of-the-art hashing approaches.

Automatic hardware/software partitioning relies on characterization, estimation and design space exploration of the system performance and cost metrics. In real world situations, such estimates are complicated and cannot be 100 percent accurate. Furthermore, hardware/software co-design is so complicated nowadays that simply considering the bipartitioning between hardware and software is not sufficient. It is important to consider some of the other key design parameters and resource sharing together with the hardware/software partitioning problem. Under variable requirements of smart systems, more flexibility on the resource usage should be incorporated in system modelling. This paper considers uncertainty modeling for system partitioning with an enhanced set of parameters for hardware/software resource sharing. We harness state-of-the-art uncertainty theory for linear uncertain distribution and normal uncertain distribution. Our derivations convert the uncertainty model back to a regular constraint optimization problem. Experimental results show the effectiveness of our approach.

Affine equivalence classification of Boolean functions has significant applications in logic synthesis and cryptography. Previous studies for classification have been limited by the large set of Boolean functions and the complex operations on the affine group. Although there are many research on affine equivalence classification for parts of Boolean functions in recent years, there are very few results for the entire set of Boolean functions. The best existing result has been achieved by Harrison with 15768919 affine equivalence classes for 6-variable Boolean functions. This paper presents a concise formula for affine equivalence classification of the entire set of Boolean functions as well as a formula for affine classification of Boolean functions with distinct ON-set size respectively. The method outlined in this paper greatly simplifies the affine group's action by constructing an isomorphism mapping from the affine group to a permutation group. By this method, we can compute the affine equivalence classes for up to 10 variables. Experiment results indicate that our scheme for calculating the affine equivalence classes for more than 6 variables is a significant advancement over previous published methods.

Network-on-chip (NoC) communication architectures present promising solutions for scalable communication requests in large system-on-chip (SoC) designs. Intellectual property (IP) core assignment and mapping are two key steps in NoC design, significantly affecting the quality of NoC systems. Both are NP-hard problems, so it is necessary to apply intelligent algorithms. In this paper, we propose improved intelligent algorithms for NoC assignment and mapping to overcome the drawbacks of traditional intelligent algorithms. The aim of our proposed algorithms is to minimize power consumption, time, area, and load balance. This work involves multiple conflicting objectives, so we combine multiple objective optimization with intelligent algorithms. In addition, we design a fault-tolerant routing algorithm and take account of reliability using comprehensive performance indices. The proposed algorithms were implemented on embedded system synthesis benchmarks suite (E3S). Experimental results show the improved algorithms achieve good performance in NoC designs, with high reliability.

Reversible quantum circuits are a necessary subclass of quantum computation and its realization is required for any quantum computer to be universal. This paper investigates how to synthesis of arbitrary ternary non-reversible logic circuits by adding inputs with constant value and garbage outputs. Group theory has been also used to solve the synthesis of reversible logic circuits. Our algorithm uses the SNT (ternary Swap gate, ternary NOT gate, ternary Toffoli gate) library, by reducing the ternary non-reversible logic circuit synthesis problem to group theory representation. The main result shows the relationship of ternary non-reversible logic circuits and the reversible circuits. The realization approach is constructive and can be further used to develop software for synthesis of arbitrary d-level circuits. This result is significantly different from the binary non-reversible logic circuits.

This paper investigates the synthesis of quantum networks built to realize hybrid switching circuits in the absence of ancilla qudits. We prove that all hybrid reversible circuits can be constructed by hybrid Not and Multiple-Controlled-Not gates. We also prove that any hybrid reversible circuit with only 1 or 2 binary qudits and arbitrary number of other qudits, can be constructed by hybrid Not and Controlled-Not gates. We present two construction-based algorithms to synthesize hybrid reversible circuits without ancilla qudits. The algorithms use hybrid Not and Multiple-Controlled-Not gates or hybrid Not and `1'-Controlled-Not gates, which are exponentially lower than breadth-first search based synthesis algorithms with respect to the input number.

Device failure is an important consideration in nano-scale design. This paper presents a probabilistic logic model to compute the probability distribution of the nano gate states. The characterization is based on markov random field and statistical physics. The basic logic gates are probabilistically characterized. The effectiveness of the method is demonstrated by an inverter and the inverter casecade. Our analysis shows that the device probability distribution highly depends on the system structures and other performance parameters.

Reversible circuits play an important role in quantum computing, which is one of the most promising emerging technologies. In this paper, we investigate the problem of optimally synthesizing 4-bit reversible circuits. We present an enhanced bi-directional synthesis approach. Owing to the exponential nature of the memory and run-time complexity, all existing methods can only perform four steps for the Controlled-Not gate NOT gate, and Peres gate library. Our novel method can achieve 12 steps. As a result, we augment the number of circuits that can optimally be synthesized by over 5 × 106 times. We synthesized 1000 random 4-bit reversible circuits. The statistical analysis result supports our estimation. The quantum cost of our result is also better than the quantum cost of other approaches. The promising experimental results demonstrate the effectiveness of our approach.

This paper presents a constructive synthesis algorithm for any n-qubit reversible function. Given any n-qubit reversible function, there are N distinct input patterns different from their corresponding outputs, where N les 2n, and the other (2n- N) input patterns will be the same as their outputs. We show that this circuit can be synthesized by at most 2nldrN '(n - 1)'-CNOT gates and 4n2ldr N NOT gates. The time complexity of our algorithm has asymptotic upper bound O(n ldr 4n). The space complexity of our synthesis algorithm is also O(n ldr 2n). The computational complexity of our synthesis algorithm is exponentially lower than the complexity of breadth-first search based synthesis algorithm.

We present a novel component-based approach to hardware/software co-verification of embedded systems using model checking. Due to their diverse applications and often strict physical constraints, embedded systems are increasingly component-based and include only the necessary components for their missions. In our approach, a component model for embedded systems which unifies the concepts of hardware IPs (i.e., hardware components) and software components is defined. Hardware and software components are verified as they are developed bottom-up. Whole systems are co-verified as they are developed top-down. Interactions of bottom-up and top-down verification are exploited to reduce verification complexity by facilitating compositional reasoning and verification reuse. Case studies on a suite of networked sensors have shown that our approach facilitates major verification reuse and leads to order-of-magnitude reduction on verification complexity

This paper proposes an approach to optimally synthesize quantum circuits by symbolic reachability analysis, where the primary inputs and outputs are basis binary and the internal signals can be nonbinary in a multiple-valued domain. The authors present an optimal synthesis method to minimize quantum cost and some speedup methods with nonoptimal quantum cost. The methods here are applicable to small reversible functions. Unlike previous works that use permutative reversible gates, a lower level library that includes nonpermutative quantum gates is used here. The proposed approach obtains the minimum cost quantum circuits for Miller gate, half adder, and full adder, which are better than previous results. This cost is minimum for any circuit using the set of quantum gates in this paper, where the control qubit of 2-qubit gates is always basis binary. In addition, the minimum quantum cost in the same manner for Fredkin, Peres, and Toffoli gates is proven. The method can also find the best conversion from an irreversible function to a reversible circuit as a byproduct of the generality of its formulation, thus synthesizing in principle arbitrary multi-output Boolean functions with quantum gate library. This paper constitutes the first successful experience of applying formal methods and satisfiability to quantum logic synthesis.

We present fast algorithms to synthesize exact minimal reversible circuits for various types of gates and costs. By reducing reversible logic synthesis problems to group theory problems, we use the powerful algebraic software GAP to solve such problems. Our algorithms are not only able to minimize for arbitrary cost functions of gates, but also orders of magnitude faster than the existing approaches to reversible logic synthesis. In addition, we show that the Peres gate is a better choice than the standard Toffoli gate in libraries of universal reversible gates.

We address the problem of implication of assertion graphs that occur in generalized symbolic trajectory evaluation (GSTE). GSTE has demonstrated its powerful capacity in formal verification of digital systems. Assertion graphs are used for property and model specifications. We present a novel implication technique for assertion graphs. It relies on direct Boolean reasoning on each edge (and vertex) of an assertion graph, thus avoiding the reachability computation in GSTE. We have successfully applied. both model-based and language-based implications on real industrial circuits. Experimental results demonstrate the promising performance of our approach.

We propose an approach to optimally synthesize quantum circuits from non-permutative quantum gates such as controlled-square-root-of-not (i.e., controlled-V). Our approach reduces the synthesis problem to multiple-valued optimization and uses group theory. We devise a novel technique that transforms the quantum logic synthesis problem from a multi-valued constrained optimization problem to a group permutation problem. The transformation enables us to utilize group theory to exploit the properties of the synthesis problem. Assuming a cost of one for each two-qubit gate, we find all reversible circuits with quantum costs of 4, 5, 6, etc, and give another algorithm to realize these reversible circuits with quantum gates.

Interconnect congestion estimation plays an important role in design automation of VLSI designs. This paper presents a novel probabilistic approach to predict the wiring space in two-dimensional arrays. We propose a hierarchical estimation method to derive approximated upper bounds for the wiring space, and we use the net density distribution to predict the routing congestion. Experimental results demonstrate the promising performance of the approach.

Reversible quantum logic plays an important role in quantum computing. In this paper, we propose an approach to optimally synthesize quantum circuits by symbolic reachability analysis where the primary inputs are purely binary. we use symbolic reachability analysis, a technique most commonly used in model checking (a way of formal verification), to synthesize the optimum quantum circuits. We present an exact synthesis method with optimal quantum cost and a speedup method with non-optimal quantum cost. Both our methods guarantee the synthesizeability of all reversible circuits. Unlike previous works which use permutative reversible gates, we use a lower level library which includes non-permutative quantum gates. For the first time, problems in quantum logic synthesis have been reduced to those of multiple-valued logic synthesis thus reducing the search space and algorithm complexity. We synthesized quantum circuits for gate, half-adder, full-adder, etc. with the smallest cost.. Our approach obtains the minimum cost quantum circuits for Miller's gate, half-adder, and full-adder, which are better than previous results. In addition, we prove the minimum quantum cost (using our elementary quantum gates) for Fredkin, Peres, and Toffoli gates. Our work constitutes the first successful experience of applying satisfiability with formal methods to quantum logic synthesis.

We present a novel symbolic routability checking approach for three-dimensional interconnect layout. The model considered is a general architecture that can fit into different applications, such as ASIC, multichip modules, field-programmable gate arrays, and reconfigurable computing architectures. The method can incrementally incorporate additional constraints driven by timing, performance, and design. We used the latest satisfiability solver to validate the effectiveness of our approach. The experimental results demonstrate the encouraging performance on difficult routing benchmarks.