Search

Supervised reconstruction models are characteristically trained on matched pairs of undersampled and fully-sampled data to capture an MRI prior, along with supervision regarding the imaging operator to enforce data consistency. To reduce supervision requirements, the recent deep image prior framework instead conjoins untrained MRI priors with the imaging operator during inference. Yet, canonical convolutional architectures are suboptimal in capturing long-range relationships, and priors based on randomly initialized networks may yield suboptimal performance. To address these limitations, here we introduce a novel unsupervised MRI reconstruction method based on zero-Shot Learned Adversarial TransformERs (SLATER). SLATER embodies a deep adversarial network with cross-attention transformers to map noise and latent variables onto coil-combined MR images. During pre-training, this unconditional network learns a high-quality MRI prior in an unsupervised generative modeling task. During inference, a zero-shot reconstruction is then performed by incorporating the imaging operator and optimizing the prior to maximize consistency to undersampled data. Comprehensive experiments on brain MRI datasets clearly demonstrate the superior performance of SLATER against state-of-the-art unsupervised methods. [ABSTRACT FROM AUTHOR]

Adversarial attacks have been demonstrated to fool the deep classification networks. There are two key characteristics of these attacks: firstly, these perturbations are mostly additive noises carefully crafted from the deep neural network itself. Secondly, the noises are added to the whole image, not considering them as the combination of multiple components from which they are made. Motivated by these observations, in this research, we first study the role of various image components and the impact of these components on the classification of the images. These manipulations do not require the knowledge of the networks and external noise to function effectively and hence have the potential to be one of the most practical options for real-world attacks. Based on the significance of the particular image components, we also propose a transferable adversarial attack against unseen deep networks. The proposed attack utilizes the projected gradient descent strategy to add the adversarial perturbation to the manipulated component image. The experiments are conducted on a wide range of networks and four databases including ImageNet and CIFAR-100. The experiments show that the proposed attack achieved better transferability and hence gives an upper hand to an attacker. On the ImageNet database, the success rate of the proposed attack is up to 88.5%, while the current state-of-the-art attack success rate on the database is 53.8%. We have further tested the resiliency of the attack against one of the most successful defenses namely adversarial training to measure its strength. The comparison with several challenging attacks shows that: (i) the proposed attack has a higher transferability rate against multiple unseen networks and (ii) it is hard to mitigate its impact. We claim that based on the understanding of the image components, the proposed research has been able to identify a newer adversarial attack unseen so far and unsolvable using the current defense mechanisms. [ABSTRACT FROM AUTHOR]

Although deep learning technologies have been widely exploited in many fields, they are vulnerable to adversarial attacks by adding small perturbations to legitimate inputs to fool targeted models. However, few studies have focused on intelligent networking in such an adversarial environment, which can pose serious security threats. In fact, while challenging intelligent networking, adversarial environments also bring about opportunities. In this paper, we, for the first time, simultaneously analyze the challenges and opportunities that the adversarial environment brings to intelligent networking. Specifically, we focus on challenges that the adversarial environment will pose on the existing intelligent networking. Furthermore, we investigate frameworks and approaches that combine adversarial machine learning with intelligent networking to solve the existing deficiencies of intelligent networking. Finally, we summarize the issues, including opportunities and challenges, which can allow researchers to focus on intelligent networking in adversarial environments. [ABSTRACT FROM AUTHOR]

• We propose to use the TA attention module to capture the relationship of different video clips. • We introduce an adversarial loss to ensure the stability and effectiveness to learn the model. • Our method achieves a competitive action evaluation result on the MTL-AQA dataset. Action quality assessment (AQA) is a task to assess the performance of a human action, which can be widely used in many real-world scenarios such as sport events. Current AQA methods generally extract features from the video and perform regression analysis to obtain the action quality score. In this process, aggregated video features may not reflect different stages of an action, which are important to evaluate an action is good or not. To address this issue, we propose to divide the video into different clips and learn the relationship between them, which may capture the action changes for accurate assessment. Time-aware (TA) attention mechanism is used to evaluate this relationship. In the experiment, our proposed method achieves promising results on the MTL-AQA dataset compared with existing AQA methods. [ABSTRACT FROM AUTHOR]

We consider optimal transport-based distributionally robust optimization (DRO) problems with locally strongly convex transport cost functions and affine decision rules. Under conventional convexity assumptions on the underlying loss function, we obtain structural results about the value function, the optimal policy, and the worst-case optimal transport adversarial model. These results expose a rich structure embedded in the DRO problem (e.g., strong convexity even if the non-DRO problem is not strongly convex, a suitable scaling of the Lagrangian for the DRO constraint, etc., which are crucial for the design of efficient algorithms). As a consequence of these results, one can develop efficient optimization procedures that have the same sample and iteration complexity as a natural non-DRO benchmark algorithm, such as stochastic gradient descent. [ABSTRACT FROM AUTHOR]

As the internet continues to be populated with new devices and emerging technologies, the attack surface grows exponentially. Technology is shifting towards a profit-driven Internet of Things market where security is an afterthought. Traditional defending approaches are no longer sufficient to detect both known and unknown attacks to high accuracy. Machine learning intrusion detection systems have proven their success in identifying unknown attacks with high precision. Nevertheless, machine learning models are also vulnerable to attacks. Adversarial examples can be used to evaluate the robustness of a designed model before it is deployed. Further, using adversarial examples is critical to creating a robust model designed for an adversarial environment. Our work evaluates both traditional machine learning and deep learning models’ robustness using the Bot-IoT dataset. Our methodology included two main approaches. First, label poisoning, used to cause incorrect classification by the model. Second, the fast gradient sign method, used to evade detection measures. The experiments demonstrated that an attacker could manipulate or circumvent detection with significant probability.

• Classifying time series with Euclidean distance is robust against adversarial attacks. • Time series classifiers using fixed kernels are robust against adversarial attacks. • This is empirically proven on 85 datasets for 2 state-of-the-art adversarial attacks. Deep neural networks have been shown to be vulnerable against specifically-crafted perturbations designed to affect their predictive performance. Such perturbations, formally termed 'adversarial attacks' have been designed for various domains in the literature, most prominently in computer vision and more recently, in time series classification. Therefore there is a need to derive robust strategies to defend deep networks from such attacks. In this work we propose to establish axioms of robustness against adversarial attacks in time series classification. We subsequently design a suitable experimental methodology and empirically validate the hypotheses put forth. Results obtained from our investigations confirm the proposed hypotheses, and provide a strong empirical baseline with a view to mitigating the effects of adversarial attacks in deep time series classification. [ABSTRACT FROM AUTHOR]

The present study, utilizing a descriptive approach and quantitative analysis, analyzed questions in the political interviews in Iran and the United States in order to show what types of adversarial are used in the journalists’ questions and whether there are differences in the use of adversarial between Iranian and American journalists. To this end, the questions addressing the presidents in Iran (i.e. Presidents AhmadiNejad and Roohani) and presidents in the US (i.e. Presidents Obama and Trump), around 70 journalists (35 in each corpus) in political press conference, were randomly collected from 2012 to 2017. The data were then analyzed based Clayman et al.’s (2006) framework to examine how language is used to express adversarial questions. The findings showed that preface tilt was significantly used more in American corpus while other-referencing frames and global adversarial were significantly used more in Iranian corpus. Moreover, in the two corpora, negative questions were the least frequently used type of question and declarative questions was absent in American corpus.

The article gives a brief historical description of the role of the court in the collection of evidence. The arguments in favor of strengthening the role of the court for the recovery of evidence are presented. In particular, the Author substantiates the conclusion that it is necessary to impose on the court the obligation to independently demand evidence in civil proceedings. The Author comes to the conclusion that strengthening the role of the court in the process of proof in civil proceedings are dictated by the requirements due to the many flaws in civil procedural law, the unequal social status of participants in the process, making decisions not in accordance with actual circumstances of the case.

Abstract Fair classification has become an important topic in machine learning research. While most bias mitigation strategies focus on neural networks, we noticed a lack of work on fair classifiers based on decision trees even though they have proven very efficient. In an up-to-date comparison of state-of-the-art classification algorithms in tabular data, tree boosting outperforms deep learning (Zhang et al. in Expert Syst Appl 82:128–150, 2017). For this reason, we have developed a novel approach of adversarial gradient tree boosting. The objective of the algorithm is to predict the output Y with gradient tree boosting while minimizing the ability of an adversarial neural network to predict the sensitive attribute S. The approach incorporates at each iteration the gradient of the neural network directly in the gradient tree boosting. We empirically assess our approach on four popular data sets and compare against state-of-the-art algorithms. The results show that our algorithm achieves a higher accuracy while obtaining the same level of fairness, as measured using a set of different common fairness definitions.

The adversarial islands that are encountered in the criminal instruction phase offer means of evidence administered in conditions of adversarial proceedings additional reliability and are an expression of the guarantees that benefit the accused person. The study is divided into three chapters, beginning with a general presentation of the adversarial principle that characterizes the trial phase and is sporadically found in the criminal investigation phase; in the following section, various concepts and theories regarding the procedures in the criminal investigation phase in which the adversarial principle is applicable are debated. In the last part of the paper, an analysis is performed in relation to the effects of the adversarial principle on the reliability of evidence obtained with its application, the exclusion of evidence obtained in violation of the adversarial principle, and lex ferenda proposals by the author. [ABSTRACT FROM AUTHOR]

In the long-term deployment of mobile robots, changing appearance brings challenges for localization. When a robot travels to the same place or restarts from an existing map, global localization is needed, where place recognition provides coarse position information. For visual sensors, changing appearances such as the transition from day to night and seasonal variation can reduce the performance of a visual place recognition system. To address this problem, we propose to learn domain-unrelated features across extreme changing appearance, where a domain denotes a specific appearance condition, such as a season or a kind of weather. We use an adversarial network with two discriminators to disentangle domain-related features and domain-unrelated features from images, and the domain-unrelated features are used as descriptors in place recognition. Provided images from different domains, our network is trained in a self-supervised manner which does not require correspondences between these domains. Besides, our feature extractors are shared among all domains, making it possible to contain more appearance without increasing model complexity. Qualitative and quantitative results on two toy cases are presented to show that our network can disentangle domain-related and domain-unrelated features from given data. Experiments on three public datasets and one proposed dataset for visual place recognition are conducted to illustrate the performance of our method compared with several typical algorithms. Besides, an ablation study is designed to validate the effectiveness of the introduced discriminators in our network. Additionally, we use a four-domain dataset to verify that the network can extend to multiple domains with one model while achieving similar performance. [ABSTRACT FROM AUTHOR]

Automatic speech emotion recognition provides computers with critical context to enable user understanding. While methods trained and tested within the same dataset have been shown successful, they often fail when applied to unseen datasets. To address this, recent work has focused on adversarial methods to find more generalized representations of emotional speech. However, many of these methods have issues converging, and only involve datasets collected in laboratory conditions. In this paper, we introduce Adversarial Discriminative Domain Generalization (ADDoG), which follows an easier to train “meet in the middle” approach. The model iteratively moves representations learned for each dataset closer to one another, improving cross-dataset generalization. We also introduce Multiclass ADDoG, or MADDoG, which is able to extend the proposed method to more than two datasets, simultaneously. Our results show consistent convergence for the introduced methods, with significantly improved results when not using labels from the target dataset. We also show how, in most cases, ADDoG and MADDoG can be used to improve upon baseline state-of-the-art methods when target dataset labels are added and in-the-wild data are considered. Even though our experiments focus on cross-corpus speech emotion, these methods could be used to remove unwanted factors of variation in other settings. [ABSTRACT FROM AUTHOR]

We introduce SPLASH units, a class of learnable activation functions shown to simultaneously improve the accuracy of deep neural networks while also improving their robustness to adversarial attacks. SPLASH units have both a simple parameterization and maintain the ability to approximate a wide range of non-linear functions. SPLASH units are: (1) continuous; (2) grounded (f (0) = 0); (3) use symmetric hinges; and (4) their hinges are placed at fixed locations which are derived from the data (i.e. no learning required). Compared to nine other learned and fixed activation functions, including ReLU and its variants, SPLASH units show superior performance across three datasets (MNIST, CIFAR-10, and CIFAR-100) and four architectures (LeNet5, All-CNN, ResNet-20, and Network-in-Network). Furthermore, we show that SPLASH units significantly increase the robustness of deep neural networks to adversarial attacks. Our experiments on both black-box and white-box adversarial attacks show that commonly-used architectures, namely LeNet5, All-CNN, Network-in-Network, and ResNet-20, can be up to 31% more robust to adversarial attacks by simply using SPLASH units instead of ReLUs. Finally, we show the benefits of using SPLASH activation functions in bigger architectures designed for non-trivial datasets such as ImageNet. [ABSTRACT FROM AUTHOR]

Deep Learning-based fault diagnosis has aroused widespread attention in machine fault diagnosis. Helicopter is an important transport for its special purpose. How to ensure its normal operation is a challenging task. Nevertheless, the existing research mainly focuses on single bearing or gear of gearbox, while there are few reports about intelligent fault diagnosis of bearing and shaft in helicopter transmission system. Furthermore, traditional domain adaption-based fault diagnosis methods assume that the source machine and target machine have the same class distribution. Besides, the latent distributed feature of target domain data is rarely developed, and the distributional discrepancy of shared-class samples during domain adaption with outlier class is rarely considered. To address these issues, we propose a domain distribution variation learning (DDVL) via adversarial adaption for helicopter transmission system fault diagnosis in this paper. Hereinto, the distributional discrepancy of partial shared-class is measured by adversarial training during open-set domain adaption. Especially, a self-supervised learning framework via the pseudo-label and weight normalization is proposed to develop the latent distribution feature of target data with unknown labels. The case study from a simulated helicopter transmission system is used to verify the effectiveness of DDVL. Our method outperforms other comparison methods in different case studies for helicopter transmission system fault diagnosis. [ABSTRACT FROM AUTHOR]

El principio de igualdad de armas es un mandato esencial que consiste en que “cada parte debe tener una oportunidad razonable para presentar su caso en condiciones que no la pongan en desventaja con respecto a su oponente”1. Esta figura, a pesar de su connotada y evidente importancia, no ha sido objeto de estudio profundo y serio en la doctrina y la jurisprudencia colombianas. Lo anterior ha permitido un uso irresponsable y equivocado de este concepto, y es de ahí que nace la necesidad de este artículo, el cual tiene por objeto, sin pretender ser exhaustivo, dar una mayor claridad sobre este mandato de op­timización, empezando por revisar sus antecedentes; pasando por el análisis de cada uno de sus elementos esenciales; la referencia a algunas categorías desarrolladas en la jurisdicción internacional que son de suma importancia para alcanzar un entendimiento íntegro de dicho principio; el señalamiento de las diferentes posiciones sobre su fundamento, definición y alcance; y una diferenciación con otros derechos y principios con los que suele confundirse. De esta manera, se sentarán las bases para que, posteriormente, sea posible abrir un debate organizado acerca de esta figura que haga viable un ejercicio optimizado y responsable de ella.

This paper presents a client/server privacy-preserving network in the context of multicentric medical image analysis. Our approach is based on adversarial learning which encodes images to obfuscate the patient identity while preserving enough information for a target task. Our novel architecture is composed of three components: 1) an encoder network which removes identity-specific features from input medical images, 2) a discriminator network that attempts to identify the subject from the encoded images, 3) a medical image analysis network which analyzes the content of the encoded images (segmentation in our case). By simultaneously fooling the discriminator and optimizing the medical analysis network, the encoder learns to remove privacy-specific features while keeping those essentials for the target task. Our approach is illustrated on the problem of segmenting brain MRI from the large-scale Parkinson Progression Marker Initiative (PPMI) dataset. Using longitudinal data from PPMI, we show that the discriminator learns to heavily distort input images while allowing for highly accurate segmentation results. Our results also demonstrate that an encoder trained on the PPMI dataset can be used for segmenting other datasets, without the need for retraining. The code is made available at: https://github.com/bachkimn/Privacy-Net-An-Adversarial-Approach-forIdentity-Obfuscated-Segmentation-of-MedicalImages [ABSTRACT FROM AUTHOR]

A smart contract honeypot is a special type of smart contract. This type of contract seems to have obvious vulnerabilities in contract design. If a user transfers a certain amount of funds to the contract, then the user can withdraw the funds in the contract. However, once users try to take advantage of this seemingly obvious vulnerability, they will fall into a real trap. Consequently, the user's investment in the contract cannot be retrieved. The honeypot induces other accounts to launch funds, which seriously threatens the security of property on the blockchain. Detection methods for honeypots are available. However, studying the manner by which to defend existing honeypots is insufficient to fight against honeypots. The new honeypots that may appear in the future from the perspective of an attacker must also be predicted. Therefore, we propose a type of adversarial honeypot. The code and behavioral features of honeypots are obtained through a comparative analysis of the 158,568 nonhoneypots and 352 honeypots. To build an adversarial honeypot, we try to separately hide these features andmake the honeypot bypass the existing detection technology. We construct 18 instances on the basis of the proposed adversarial honeypot and use an open-source honeypot detection tool to detect these instances. The experimental result shows that the proposed honeypot can bypass the detection tool with a 100% ratio. Therefore, this type of honeypot should be given attention, and defensive measures should be proposed as soon as possible. [ABSTRACT FROM AUTHOR]

In the article, the author deals with the issue of adversariality of criminal proceedings as one of the basic principles of criminal proceedings with a focus on the application of this principle in criminal proceedings. In the article, the author focused on clarifying the concept of adversariality, and followed up on the general definition of adversariality as a principle of criminal procedure. Given that this is a category of broad importance, the author tried to define these definitions as precisely as possible and did not focus on its comparison with other principles of criminal proceedings, but focused only on the principle of adversariality. Furthermore, in the article, the author focused on the application of the adversarial principle in selected post-clerical acts of criminal proceedings with a focus on the exercise of the rights of the accused and exceptions to the principle of adversariality in preparatory proceedings. In the article, the author deals with the issue of adversariality of criminal proceedings as one of the basic principles of criminal proceedings with a focus on the application of this principle in criminal proceedings. In the article, the author focused on clarifying the concept of adversariality, and followed up on the general definition of adversariality as a principle of criminal procedure. Given that this is a category of broad meaning, the author tried to define these definitions as precisely as possible and did not focus on its comparison with other principles of criminal proceed-ings, but focused only on the principle of adversariality. Furthermore, in the article, the author focused on the applica-tion of the adversarial principle in selected post-clerical acts of criminal proceedings with a focus on the exercise of the rights of the accused and exceptions to the principle of adversariality in preparatory proceedings. Attention is also paid to the violation of the principle of contradictory in the preparatory proceedings in the fourth chapter. [ABSTRACT FROM AUTHOR]

Explaining cases of stopping and dropping the litigation and their reasons in the Jordanian Code of Procedure, the problematic of this study is to explain the stopping of the Jordanian legislator from cases of stopping and dropping the litigation and the adequacy of the legal texts regulating it in Jordanian legislation. The two researchers followed the descriptive and analytical approach in analyzing legal texts, particularly the texts of the Jordanian Code of Civil Procedure, the jurisprudence, and the opinions of the jurists. Likewise, the comparative approach, whenever necessary, in order to enrich the study, and the researchers reached a number of findings and recommendations, and the most important result was: that the litigation may be intercepted by situations that lead to affecting the course of its outcome, such as endowment and omission. The researchers recommended the necessity of expanding the study of the judicial litigation and what influences it in the cases of stopping and abrogation, due to its importance in determining the legal positions of the parties. [ABSTRACT FROM AUTHOR]

Recent studies have shown that state-of-the-art DNNs are not always credible, despite their impressive performance on the hold-out test set of a variety of tasks. These models tend to exploit dataset shortcuts to make predictions, rather than learn the underlying task. The non-credibility could lead to low generalization, adversarial vulnerability, as well as algorithmic discrimination of the DNN models. In this paper, we propose CREX in order to develop more credible DNNs. The high-level idea of CREX is to encourage DNN models to focus more on evidences that actually matter for the task at hand and to avoid overfitting to data-dependent shortcuts. Specifically, in the DNN training process, CREX directly regularizes the local explanation with expert rationales, i.e., a subset of features highlighted by domain experts as justifications for predictions, to enforce the alignment between local explanations and rationales. Even when rationales are not available, CREX still could be useful by requiring the generated explanations to be sparse. In addition, CREX is widely applicable to different network architectures, including CNN, LSTM and attention model. Experimental results on several text classification datasets demonstrate that CREX could increase the credibility of DNNs. Comprehensive analysis further shows three meaningful improvements of CREX: (1) it significantly increases DNN accuracy on new and previously unseen data beyond test set, (2) it enhances fairness of DNNs in terms of equality of opportunity metric and reduce models' discrimination toward certain demographic group, and (3) it promotes the robustness of DNN models with respect to adversarial attack. These experimental results highlight the advantages of the increased credibility by CREX. [ABSTRACT FROM AUTHOR]

We derive an algorithm that achieves the optimal (within constants) pseudo-regret in both adversarial and stochastic multi-armed bandits without prior knowledge of the regime and time horizon.1 The algorithm is based on online mirror descent (OMD) with Tsallis entropy regularization with power = 1=2 and reduced-variance loss estimators. More generally, we define an adversarial regime with a self-bounding constraint, which includes stochastic regime, stochastically constrained adversarial regime (Wei and Luo, 2018), and stochastic regime with adversarial corruptions (Lykouris et al., 2018) as special cases, and show that the algorithm achieves logarithmic regret guarantee in this regime and all of its special cases simultaneously with the optimal regret guarantee in the adversarial regime. The algorithm also achieves adversarial and stochastic optimality in the utility-based dueling bandit setting. We provide empirical evaluation of the algorithm demonstrating that it significantly outperforms Ucb1 and Exp3 in stochastic environments. We also provide examples of adversarial environments, where Ucb1 and Thompson Sampling exhibit almost linear regret, whereas our algorithm suers only logarithmic regret. To the best of our knowledge, this is the first example demonstrating vulnerability of Thompson Sampling in adversarial environments. Last but not least, we present a general stochastic analysis and a general adversarial analysis of OMD algorithms with Tsallis entropy regularization for 2 [0; 1] and explain the reason why = 1=2 works best. [ABSTRACT FROM AUTHOR]

• A modality invariant adversarial mechanism for improving the multi-style Re-ID task. • Two new datasets from virtuality to reality for the multi-style Re-ID task. • Space transformation and different category classifiers for performance improvement. Person re-identification based on multi-style images helps in crime scene investigation, where only a virtual image (sketch or portrait) of the suspect is available for retrieving possible identities. However, due to the modality gap between multi-style images, standard model of person re-identification cannot achieve satisfactory performance when directly applied to match the virtual images with the real photographs. To address this problem, we propose a modality invariant adversarial mechanism (MIAM) to remove the modality gap between multi-style images. Specifically, the MIAN consists of two parts: a space transformation module to transfer the multi-style person images to a modality-invariant space, and an adversarial learning module "played" between the category classifier and modality classifier to steer the representation learning. The modality classifier discriminates between the real and virtual images while the category classifier predicts the identities of the input transformed images. We explore the space transformation for data augmentation to further bridge the modality gap and facilitate the performance. Furthermore, we build two new datasets for the multi-style Re-ID to evaluate the performance. Extensive experimental results demonstrate the effectiveness of the proposed method on improving the performance against the existing feature learning networks. Further comparison results conducted on different modules in MIAM show that our approach is of favorable generalization ability on alleviating the modality gap to improve the multi-style Re-ID. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of semi-supervised transfer learning with limited cross-modality data in remote sensing. A large amount of multi-modal earth observation images, such as multispectral imagery (MSI) or synthetic aperture radar (SAR) data, are openly available on a global scale, enabling parsing global urban scenes through remote sensing imagery. However, their ability in identifying materials (pixel-wise classification) remains limited, due to the noisy collection environment and poor discriminative information as well as limited number of well-annotated training images. To this end, we propose a novel cross-modal deep-learning framework, called X-ModalNet, with three well-designed modules: self-adversarial module, interactive learning module, and label propagation module, by learning to transfer more discriminative information from a small-scale hyperspectral image (HSI) into the classification task using a large-scale MSI or SAR data. Significantly, X-ModalNet generalizes well, owing to propagating labels on an updatable graph constructed by high-level features on the top of the network, yielding semi-supervised cross-modality learning. We evaluate X-ModalNet on two multi-modal remote sensing datasets (HSI-MSI and HSI-SAR) and achieve a significant improvement in comparison with several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Owing to the advantages of low storage cost and high query efficiency, cross-modal hashing has received increasing attention recently. As failing to bridge the inherent modality gap between modalities, most existing cross-modal hashing methods have limited capability to explore the semantic consistency information between different modality data, leading to unsatisfactory search performance. To address this problem, we propose a novel deep hashing method named Multi-Task Consistency-Preserving Adversarial Hashing (CPAH) to fully explore the semantic consistency and correlation between different modalities for efficient cross-modal retrieval. First, we design a consistency refined module (CR) to divide the representations of different modality into two irrelevant parts, i.e., modality-common and modality-private representations. Then, a multi-task adversarial learning module (MA) is presented, which can make the modality-common representation of different modalities close to each other on feature distribution and semantic consistency. Finally, the compact and powerful hash codes can be generated from modality-common representation. Comprehensive evaluations conducted on three representative cross-modal benchmark datasets illustrate our method is superior to the state-of-the-art cross-modal hashing methods. [ABSTRACT FROM AUTHOR]

Fair classification has become an important topic in machine learning research. While most bias mitigation strategies focus on neural networks, we noticed a lack of work on fair classifiers based on decision trees even though they have proven very efficient. In an up-to-date comparison of state-of-the-art classification algorithms in tabular data, tree boosting outperforms deep learning (Zhang et al. in Expert Syst Appl 82:128–150, 2017). For this reason, we have developed a novel approach of adversarial gradient tree boosting. The objective of the algorithm is to predict the output Y with gradient tree boosting while minimizing the ability of an adversarial neural network to predict the sensitive attribute S. The approach incorporates at each iteration the gradient of the neural network directly in the gradient tree boosting. We empirically assess our approach on four popular data sets and compare against state-of-the-art algorithms. The results show that our algorithm achieves a higher accuracy while obtaining the same level of fairness, as measured using a set of different common fairness definitions. [ABSTRACT FROM AUTHOR]

Compared to facial expression recognition, expression synthesis requires a very high-dimensional mapping. This problem exacerbates with increasing image sizes and limits existing expression synthesis approaches to relatively small images. We observe that facial expressions often constitute sparsely distributed and locally correlated changes from one expression to another. By exploiting this observation, the number of parameters in an expression synthesis model can be significantly reduced. Therefore, we propose a constrained version of ridge regression that exploits the local and sparse structure of facial expressions. We consider this model as masked regression for learning local receptive fields. In contrast to the existing approaches, our proposed model can be efficiently trained on larger image sizes. Experiments using three publicly available datasets demonstrate that our model is significantly better than ℓ 0 , ℓ 1 and ℓ 2 -regression, SVD based approaches, and kernelized regression in terms of mean-squared-error, visual quality as well as computational and spatial complexities. The reduction in the number of parameters allows our method to generalize better even after training on smaller datasets. The proposed algorithm is also compared with state-of-the-art GANs including Pix2Pix, CycleGAN, StarGAN and GANimation. These GANs produce photo-realistic results as long as the testing and the training distributions are similar. In contrast, our results demonstrate significant generalization of the proposed algorithm over out-of-dataset human photographs, pencil sketches and even animal faces. [ABSTRACT FROM AUTHOR]

The wrongful conviction of David Harold Eastman in the Australian Capital Territory represents one of Australia's most recent and high-profile public failures of the criminal justice system and highlights the limits of the Australian legal system. Further, the Eastman case draws into question the use of inquiries into miscarriages of justice, particularly when an inquiry's recommendations can be disregarded by governments (as it was in this instance). This article provides an overview of the Eastman case and critically evaluates how it sheds light on the use of inquiries as an avenue to investigate and correct wrongful convictions more broadly in Australia. [ABSTRACT FROM AUTHOR]

Semantic representations in higher sensory cortices form the basis for robust, yet flexible behavior. These representations are acquired over the course of development in an unsupervised fashion and continuously maintained over an organism's lifespan. Predictive processing theories propose that these representations emerge from predicting or reconstructing sensory inputs. However, brains are known to generate virtual experiences, such as during imagination and dreaming, that go beyond previously experienced inputs. Here, we suggest that virtual experiences may be just as relevant as actual sensory inputs in shaping cortical representations. In particular, we discuss two complementary learning principles that organize representations through the generation of virtual experiences. First, "adversarial dreaming" proposes that creative dreams support a cortical implementation of adversarial learning in which feedback and feedforward pathways engage in a productive game of trying to fool each other. Second, "contrastive dreaming" proposes that the invariance of neuronal representations to irrelevant factors of variation is acquired by trying to map similar virtual experiences together via a contrastive learning process. These principles are compatible with known cortical structure and dynamics and the phenomenology of sleep thus providing promising directions to explain cortical learning beyond the classical predictive processing paradigm. [ABSTRACT FROM AUTHOR]

Artificial Intelligence (AI) is one of the hottest topics in our society, especially when it comes to solving data-analysis problems. Industry are conducting their digital shifts, and AI is becoming a cornerstone technology for making decisions out of the huge amount of (sensors-based) data available in the production floor. However, such technology may be disappointing when deployed in real conditions. Despite good theoretical performances and high accuracy when trained and tested in isolation, a Machine-Learning (M-L) model may provide degraded performances in real conditions. One reason may be fragility in treating properly unexpected or perturbed data. The objective of the paper is therefore to study the robustness of seven M-L and Deep-Learning (D-L) algorithms, when classifying univariate time-series under perturbations. A systematic approach is proposed for artificially injecting perturbations in the data and for evaluating the robustness of the models. This approach focuses on two perturbations that are likely to happen during data collection. Our experimental study, conducted on twenty sensors’ datasets from the public University of California Riverside (UCR) repository, shows a great disparity of the models’ robustness under data quality degradation. Those results are used to analyse whether the impact of such robustness can be predictable—thanks to decision trees—which would prevent us from testing all perturbations scenarios. Our study shows that building such a predictor is not straightforward and suggests that such a systematic approach needs to be used for evaluating AI models’ robustness.

El presente documento tiene como propósito efectuar una aproximación, desde el punto de vista jurisprudencial y legal, del desarrollo que ha tenido la prueba oficiosa en el sistema penal adversarial de tendencia acusatoria, implementado en Colombia con la Ley 906 de 2004, a fin de establecer la tensión existente entre los principios de imparcialidad, igualdad de armas y legalidad, frente al principio de justicia material y en el contexto del rol del juez en el proceso. En efecto, esta es una de las discusiones que enfrentan quienes administran justicia y quienes acuden a ella para resolver los conflictos sociales, en la que se evidencia que la construcción de un sistema acusatorio es un proceso inacabado y en constante evolución. Para ello se hace necesario abordar no solo un concepto meramente legalista, sino que se requiere recabar en la filosofía misma del sistema penal acusatorio y las realidades en que este se aplica, para cumplir los cometidos sociales para los cuales fue creado.