This article addresses the need for adaptive ethical analysis within machine learning that accounts for emerging problems concerning social bias and generative adversarial networks. I use John Dewey’s criticisms of the reflex arc concept in psychology as a basis for understanding how these problems stem from human-gan interaction. By combining Dewey’s criticisms with Donna Haraway’s idea of cyborgs, Luciano Floridi’s concept of distributed morality, and Shaowen Bardzell’s recommendations for a feminist approach to human-computer interaction, I suggest a (...) dynamic perspective from which to begin analyzing and solving issues of injustice evident in this particular domain of machine learning. (shrink)

Current development in a deep neural network has given an opportunity to a novel framework for the reconstruction of a holographic image and a phase recovery method with real-time performance. There are many deep learning-based techniques that have been proposed for the holographic image reconstruction, but these deep learning-based methods can still lack in performance, time complexity, accuracy, and real-time performance. Due to iterative calculation, the generation of a CGH requires a long computation time. A novel deep generative (...) class='Hi'>adversarial network holography framework is proposed for hologram reconstruction. This novel framework consists of two phases. In phase one, we used the Fresnel-based method to make the dataset. In the second phase, we trained the raw input image and holographic label image data from phase one acquired images. Our method has the capability of the noniterative process of computer-generated holograms. The experimental results have demonstrated that the proposed method outperforms the existing methods. (shrink)

Real-time smoke detection is of great significance for early warning of fire, which can avoid the serious loss caused by fire. Detecting smoke in actual scenes is still a challenging task due to large variance of smoke color, texture, and shapes. Moreover, the smoke detection in the actual scene is faced with the difficulties in data collection and insufficient smoke datasets, and the smoke morphology is susceptible to environmental influences. To improve the performance of smoke detection and solve the problem (...) of too few datasets in real scenes, this paper proposes a model that combines a deep convolutional generative adversarial network and a convolutional neural network to extract smoke features and detection. The vibe algorithm was used to collect smoke and nonsmoke images in the dynamic scene and deep convolutional generative adversarial network used these images to generate images that are as realistic as possible. Besides, we designed an improved convolutional neural network model for extracting smoke features and smoke detection. The experimental results show that the method has a good detection performance on the smoke generated in the actual scenes and effectively reduces the false alarm rate. (shrink)

The scarcity of abnormal data leads to imbalanced data in the field of monitoring tool wear conditions. In this paper, a novel multi-index generative adversarial network is proposed to detect the tool wear conditions subject to imbalanced signal data. First, the generator in the MI-GAN is trained to produce fake normal signals, and the discriminator computes scores of testing signals and generated signals. Next, the generator detects abnormal signals based on the performance of imitating testing signals, and the (...) discriminator will compute the scores of testing signals and generated signals. Subsequently, two indexes, i.e., L 2 -norm and temporal correlation coefficient, are put forward to measure the similarity between generated signals and testing signals. Finally, our decision-making function further combines L 2 -norm and CORT with two discriminator scores to determine the tool conditions. Experimental results show that our method obtains 97% accuracy in tool wear detection based on imbalanced data without manual feature extraction, which outperforms traditional machine learning methods. (shrink)

The text analyses the new images produced by artificial neural networks such as Generative Adversarial Networks (GANs) from the perspective of photography and, more specifically, cameraless photography. The images produced by GANs are located within the wider framework of the impact of machine learning technologies on contemporary visual culture and contemporary artistic practices. In the final section, the article focuses on the work of two artists who have explicity tackled the relations between GAN-generated images and the (...) traditions of photography and cameraless photography, with their multiple intertwinings of human and non-human agencies: Mario Klingemann and Grégory Chatonsky. (shrink)

In this paper, I describe my artistic project, Igùn—a StyleGAN series trained to animate the research question: what bronze objects could have been produced should the 1897 British invasion not have occurred in the Benin Kingdom? In addition to looting over 3000 palace-commissioned artworks, I surmise that the invasion resulted in a 17-year (1897–1914) artistic decline. Although post-invasion colonial reports referred to a thriving art scene and increased colonial art patronage, there is a dearth of visual documentation to identify objects (...) created during this period. Considering this absence, I propose Igùn, a series of StyleGAN models trained on a dataset of looted Benin Bronzes. This project is informed by the Igún Eronmwon’s (the royal guild of bronze casters) artistic protocols. Finally, I present three prototypes based on emergent themes—infancy and facial expressions, which were underexplored in Benin’s classical bronze casting tradition. (shrink)

The generation of synthetic data is becoming a fundamental task in the daily life of any organization due to the new protection data laws that are emerging. Because of the rise in the use of Artificial Intelligence, one of the most recent proposals to address this problem is the use of Generative Adversarial Networks. These types of networks have demonstrated a great capacity to create synthetic data with very good performance. The goal of synthetic data generation (...) is to create data that will perform similarly to the original dataset for many analysis tasks, such as classification. The problem of GANs is that in a classification problem, GANs do not take class labels into account when generating new data, it is treated as any other attribute. This research work has focused on the creation of new synthetic data from datasets with different characteristics with a Conditional Generative Adversarial Network. CGANs are an extension of GANs where the class label is taken into account when the new data is generated. The performance of our results has been measured in two different ways: firstly, by comparing the results obtained with classification algorithms, both in the original datasets and in the data generated; secondly, by checking that the correlation between the original data and those generated is minimal. (shrink)

Until now, the mass spread of fake news and its negative consequences have implied mainly textual content towards a loss of citizens' trust in institutions. Recently, a new type of machine learning framework has arisen, Generative Adversarial Networks (GANs) – a class of deep neural network models capable of creating multimedia content (photos, videos, audio) that simulate accurate content with extreme precision. While there are several areas of worthwhile application of GANs – e.g., in the field of (...) audio-visual production, human-computer interactions, satire, and artistic creativity – their deceptive uses, at least as currently foreseeable, are just as numerous and worrying. The main concern is linked to the so-called “deepfakes”, fake images or videos that simulate real events with extreme precision. When trained on a human face, GANs can make the face assume hyper-realistic movements, expressions and (verbal and non-verbal) communication abilities. This technology poses an urgent threat to the governance of democratic processes concerning the production of public opinions and political discourses, with significant potential for reality-altering and disinformation. After a short introduction of their current technical state-of-the-art, in this paper we want to enquire the GANs` socio-technical system alongside different and intertwined philosophical accounts. Firstly, we will argue about the conditions that make perceived a GANs-generated content as trustworthy, arguing also about the general effects GANs might have on the perceived trustworthiness of individuals. Thereafter, we will discuss about the inadequacy to approach GANs only as perception-altering technology. Against this backdrop, we will propose a theoretical turn that considers the human-machine relationships of trustworthiness as elements of a broader hybrid socio-technical systems. This turn come up with political repercussions that we will discuss in the last part of the paper. (shrink)

The ongoing coronavirus 2019 pandemic caused by the severe acute respiratory syndrome coronavirus 2 has resulted in a severe ramification on the global healthcare system, principally because of its easy transmission and the extended period of the virus survival on contaminated surfaces. With the advances in computer-aided diagnosis and artificial intelligence, this paper presents the application of deep learning and adversarial network for the automatic identification of COVID-19 pneumonia in computed tomography scans of the lungs. The complexity and time (...) limitation of the reverse transcription-polymerase chain reaction swab test makes it disadvantageous to depend solely on as COVID-19’s central diagnostic mechanism. Since CT imaging systems are of low cost and widely available, we demonstrate that the drawback of the RT-PCR can be alleviated with a faster, automated, and reduced contact diagnostic process via the use of a neural network model for the classification of infected and noninfected CT scans. In our proposed model, we explore the benefit of transfer learning as a means of resolving the problem of inadequate dataset and the importance of semisupervised generative adversarial network for the extraction of well-mapped features and generation of image data. Our experimental evaluation indicates that the proposed semisupervised model achieves reliable classification, taking advantage of the reflective loss distance between the real data sample space and the generated data. (shrink)

This paper provides a comprehensive exploration of the transformative role of generative AI models, specifically Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), in the realm of medical diagnosis. Drawing from the philosophy of medicine and epidemiology, the paper examines the technical, ethical, and philosophical dimensions of integrating generative models into healthcare. A case study featuring Emily underscores the pivotal support generative AI can offer in complex medical diagnoses. The discussion extends to the application (...) of GANs and VAEs in medical imaging, emphasizing their potential in improving diagnostics, treatment planning, and medical research. The paper further delves into challenges and controversies, addressing issues of anatomical accuracy, biases in training data, interpretability of AI-generated medical images, and ethical considerations, for example, the phenomenon of "Dr. Google" and its implications for self-diagnosis, particularly in the context of the increasing role of generative AI models in healthcare. The concluding section emphasizes the need for health literacy, responsible use of online information, and collaborative decision-making between patients and healthcare providers. We advocate for interdisciplinary collaborations to establish ethical guidelines and ensure responsible AI use in healthcare. (shrink)

Comparatively easy questions we might ask about creativity are distinguished from the hard question of explaining transformative creativity. Many have focused on the easy questions, offering no reason to think that the imagining relied upon in creative cognition cannot be reduced to more basic folk psychological states. The relevance of associative thought processes to songwriting is then explored as a means for understanding the nature of transformative creativity. Productive artificial neural networks—known as generative antagonistic networks (GANs)—are a (...) recent example of how a system’s ability to generate novel products can both be finely tuned by prior experience and grounded in strategies that cannot be articulated by the system itself. Further, the kinds of processes exploited by GANs need not be seen as incorporating something akin to sui generis imaginative states. The chapter concludes with reflection on the added relevance of personal character to explanations of creativity. [This is Chapter 12 of the book Explaining Imagination.]. (shrink)

Digitization of every daily procedure requires trustworthy verification schemes. People tend to overlook the security of the passwords they use, i.e. they use the same password on different occasions, they neglect to change them periodically or they often forget them. This raises a major security issue, especially for elderly people who are not familiar with modern technology and its risks and challenges. To overcome these drawbacks, biometric factors were utilized, and nowadays, they have been widely adopted due to their convenience (...) of use and ease of hardware installation. Many biometric-based authentication schemes were proposed, but despite the advantages that they offer, recent research has shown that biometrics by itself cannot be considered as an inviolable technique. Recently, we have proposed StegoPass, a novel method that obtains the 68 facial points of a user and utilizes them as a stego message to an image. This produced stego image was the password. Although the experiments conducted showed maximum security, it would be challenging to enhance the robustness of the proposed model for even more attacks. This paper examines the utilization of multimodal biometrics as the secret message embedded in the image. More specifically, besides the extraction of the facial points, we extract the unique minutiae moments and combine them in a feature vector. This feature vector is then embedded in the image. Two different datasets were used, and the security of the method was tested against state-of-the-art deep learning models, i.e. generative adversarial networks, to test whether the image could be digitally synthesized and fool the proposed verification scheme. The results proved that the new enhanced version of StegoPass offers an extremely secure method as its predecessor. (shrink)

In the last 20 years the Turing test has been left further behind by new developments in artificial intelligence. At the same time, however, these developments have revived some key elements of the Turing test: imitation and adversarialness. On the one hand, many generative models, such as generative adversarial networks, build imitators under an adversarial setting that strongly resembles the Turing test. The term “Turing learning” has been used for this kind of setting. On the (...) other hand, AI benchmarks are suffering an adversarial situation too, with a ‘challenge-solve-and-replace’ evaluation dynamics whenever human performance is ‘imitated’. The particular AI community rushes to replace the old benchmark by a more challenging benchmark, one for which human performance would still be beyond AI. These two phenomena related to the Turing test are sufficiently distinctive, important and general for a detailed analysis. This is the main goal of this paper. After recognising the abyss that appears beyond superhuman performance, we build on Turing learning to identify two different evaluation schemas: Turing testing and adversarial testing. We revisit some of the key questions surrounding the Turing test, such as ‘understanding’, commonsense reasoning and extracting meaning from the world, and explore how the new testing paradigms should work to unmask the limitations of current and future AI. Finally, we discuss how behavioural similarity metrics could be used to create taxonomies for artificial and natural intelligence. Both testing schemas should complete a transition in which humans should give way to machines—not only as references to be imitated but also as judges—when pursuing and measuring machine intelligence. (shrink)

‘I am not data’ has been produced in collaboration with a creative coder from the Netherlands. Over a period of six months around 4600 images were chosen from approximately 20,000 images of LatinX femmes. The images were then fed into a generative adversarial network (GAN) simulation that produced this film. The resulting work seeks to elucidate a process of lumping together foreign bodies as a method of amalgamation that works to create ‘new’ information. It exists in tandem with (...) ‘Second Nature’, a body of work on Mexican femme bodies and algorithmic web searches, from which some images are included in this article. The process behind the GAN used for this film becomes a mere reference to a human shape. Its processes of synthesis work as a metaphor for systems of human classification in which the bodies of foreigners become fuel for monstrous stereotypes. There is a sense of history repeating itself through the ways that these technological tools are used in artificial intelligence (AI) which launches questions regarding equality and progress in pursuit of social justice. (shrink)

One of the best known arguments against the connectionist approach to artificial intelligence and cognitive science is that neural networks are black boxes, i.e., there is no understandable account of their operation. This difficulty has impeded efforts to explain how categories arise from raw sensory data. Moreover, it has complicated investigation about the role of symbols and language in cognition. This state of things has been radically changed by recent experimental findings in artificial deep learning research. Two kinds of (...) artificial deep learning networks, namely the convolutional neural network and the generative adversarial network have been found to possess the capability to build internal states that are interpreted by humans as complex visual categories, without any specific hints or any grammatical processing. This emergent ability suggests that those categories do not depend on human knowledge or the syntactic structure of language, while they do rely on their visual context. This supports a mild form of empiricism, while it does not assume that computational functionalism is true. Some consequences are extracted regarding the debate about amodal and grounded representations in the human brain. Furthermore, new avenues for research on cognitive science are open. (shrink)

Current deep neural networks (DNNs) are far from being able to model the rich landscape of human visual experience. Beyond visual recognition, we explore the neural substrates of visual mental imagery and other visual experiences. Rather than shared visual representations, temporal dynamics and functional connectivity of the process are essential. Generative adversarial networks may drive future developments in simulating human visual experience.

This paper explores the intersection of art history and AI technology. Special attention is paid to Generative Adversarial Networks (GANs), a machine learning technology widely used in AI art. This technology is particularly interesting to art history and the philosophy of art because it raises enduring questions about the creative process of artmaking, especially what constitutes a new and original work of art. While this is a relatively new area, it is possible to discern emerging directions where (...) art and AI intersect. One pressing issue concerns the use (or misuse) of algorithmically enabled technologies and datasets to perpetuate longstanding art-historical biases. The critical position adopted in this paper shares this concern. I examine a notable example of GAN technology that claims to create original art styles. While such experiments in computational creativity are technically interesting, their artistic aims can be questionable. I argue a mechanical model of artistic development based solely on the autonomy of style is problematic for at least two reasons: 1) it uncritically reinstates a formalist view of art history; 2) it recreates a narrow style-centric ‘historical modernism.’ However, despite voicing these reservations, I also discuss the potential of this technology to engage with topics of critical relevance to art history and the philosophy of art. (shrink)

Artificial intelligence (AI) is rapidly changing the way that art is created and consumed, allowing artists to create unique, engaging works with high computing power that can supplement their creative process. This manuscript explores the creative process of using AI technology in digital art to create paintings and evaluates creativity based on the aesthetic value and components of works created by AI. This research seeks to understand how AI technology influences the art world through a practice-led methodology with a descriptive (...) qualitative approach and observational methodology. Pioneering artwork such as Edmond de Belamy, created by French collective obvious using generative adversarial networks (GANs), has raised questions among the creative fields concerning originality, creativity, and authorship. Through analyses of AI-generated artwork and comparison with traditional artwork, the research investigates potential opportunities and challenges that AI presents to the art world. Ultimately, AI technology can provide a creative spark for human artists who, with an appropriate understanding of the limitations of AI algorithms, can create incredible pieces of art that embrace the contributions of both humans and machines. This research seeks to fill in the gaps left by earlier works and provide a more multi-faceted look at the changing landscape of AI in art, gaining a better understanding of the ways AI is impacting the art world and what the future may look like in the coming years. (shrink)

With the expansion of people’s needs, the translation performance of traditional models is increasingly unable to meet current demands. This article mainly studied the Transformer model. First, the structure and principle of the Transformer model were briefly introduced. Then, the model was improved by a generative adversarial network to improve the translation effect of the model. Finally, experiments were carried out on the linguistic data consortium dataset. It was found that the average Bilingual Evaluation Understudy value of the (...) improved Transformer model improved by 0.49, and the average perplexity value reduced by 10.06 compared with the Transformer model, but the computation speed was not greatly affected. The translation results of the two example sentences showed that the translation of the improved Transformer model was closer to the results of human translation. The experimental results verify that the improved Transformer model can improve the translation quality and be further promoted and applied in practice to further improve the English translation and meet application needs in real life. (shrink)

Aiming at the problem of radar target recognition of High-Resolution Range Profile under low signal-to-noise ratio conditions, a recognition method based on the Constrained Naive Least-Squares Generative Adversarial Network, Short-time Fourier Transform, and Convolutional Neural Network is proposed. Combining the Least-Squares Generative Adversarial Network with the Wasserstein Generative Adversarial Network with Gradient Penalty, the CN-LSGAN is presented and applied to the HRRP denoise. The frequency domain and phase features of HRRP are gained by STFT (...) in order to facilitate feature learning and also match the input data format of the CNN. These experimental results show that the CN-LSGAN has better data augmentation performance and can effectively avoid the model collapse compared to the generative adversarial network and LSGAN. Also, the method has better recognition performance than the one-dimensional CNN method and the Long Short-Term Memory network method. (shrink)

Computational visual perception, also known as computer vision, is a field of artificial intelligence that enables computers to process digital images and videos in a similar way as biological vision does. It involves methods to be developed to replicate the capabilities of biological vision. The computer vision’s goal is to surpass the capabilities of biological vision in extracting useful information from visual data. The massive data generated today is one of the driving factors for the tremendous growth of computer vision. (...) This survey incorporates an overview of existing applications of deep learning in computational visual perception. The survey explores various deep learning techniques adapted to solve computer vision problems using deep convolutional neural networks and deep generative adversarial networks. The pitfalls of deep learning and their solutions are briefly discussed. The solutions discussed were dropout and augmentation. The results show that there is a significant improvement in the accuracy using dropout and data augmentation. Deep convolutional neural networks’ applications, namely, image classification, localization and detection, document analysis, and speech recognition, are discussed in detail. In-depth analysis of deep generative adversarial network applications, namely, image-to-image translation, image denoising, face aging, and facial attribute editing, is done. The deep generative adversarial network is unsupervised learning, but adding a certain number of labels in practical applications can improve its generating ability. However, it is challenging to acquire many data labels, but a small number of data labels can be acquired. Therefore, combining semisupervised learning and generative adversarial networks is one of the future directions. This article surveys the recent developments in this direction and provides a critical review of the related significant aspects, investigates the current opportunities and future challenges in all the emerging domains, and discusses the current opportunities in many emerging fields such as handwriting recognition, semantic mapping, webcam-based eye trackers, lumen center detection, query-by-string word, intermittently closed and open lakes and lagoons, and landslides. (shrink)

Grammatical error correction is an important element in language learning. In this article, based on deep learning, the application of the Transformer model in GEC was briefly introduced. Then, in order to improve the performance of the model on GEC, it was optimized by a generative adversarial network. Experiments were conducted on two data sets. It was found that the performance of the GAN-combined Transformer model was significantly improved compared to the Transformer model. The F 0.5 value of (...) the optimized model was 53.87 on CoNIL-2014, which was 2.69 larger than the Transformer model; the generalized language evaluation understanding value of the optimized model was 61.77 on JFLEG, which was 8.81 larger than that of the Transformer model. The optimized model also had a favorable correction performance in an actual English essay. The experimental results verify the reliability of the GAN-combined Transformer model on automatic English GEC, suggesting that the model can be further promoted and applied in practice. (shrink)

Borrowing Timothy Morton’s notion of hyperobject, this article explores questions of network and scale in generative adversarial networks (GAN) images. In this context, the term network refers to the omnipresence of algorithmic images today and their significant impact on our lives. Such images are massively distributed in time and space beyond any sensible human-scale. Scale, in this context, denotes the relations between different operational layers of algorithmic images, such as the pictorial layer in contrast to the data (...) layer. An algorithmic image is simultaneously a visual image, a symbol, a data point and part of a mass visual milieu. Its meaning is thus polymorphic and can, arguably, never be exhausted. The article explores these terms through analysis of the website www.thispersondoesnotexist.com. (shrink)

Imagine a world where machines can see and understand the world the way humans do. Rapid progress in artificial intelligence has led to smartphones that recognize faces, cars that detect pedestrians, and algorithms that suggest diagnoses from clinical images, among many other applications. The success of computer vision is founded on a deep understanding of the neural circuits in the brain responsible for visual processing. This book introduces the neuroscientific study of neuronal computations in visual cortex alongside of the psychological (...) understanding of visual cognition and the burgeoning field of biologically-inspired artificial intelligence. Topics include the neurophysiological investigation of visual cortex, visual illusions, visual disorders, deep convolutional neural networks, machine learning, and generative adversarial networks among others. It is an ideal resource for students and researchers looking to build bridges across different approaches to studying and developing visual systems. (shrink)

The making of infrared templates is of great significance for improving the accuracy and precision of infrared imaging guidance. However, collecting infrared images from fields is difficult, of high cost, and time-consuming. In order to address this problem, an infrared image generation method, infrared generative adversarial networks, based on conditional generative adversarial networks architecture is proposed. In I-GANs, visible images instead of random noise are used as the inputs, and the D-LinkNet network is also (...) utilized to build the generative model, enabling improved learning of rich image textures and identification of dependencies between images. Moreover, the PatchGAN architecture is employed to build a discriminant model to process the high-frequency components of the images effectively and reduce the amount of calculation required. In addition, batch normalization is used to optimize the training process, and thereby, the instability and mode collapse of the generated adversarial network training can be alleviated. Finally, experimental verification is conducted on the produced infrared/visible light dataset. The experimental results reveal that high-quality and reliable infrared data are generated by the proposed I-GANs. (shrink)

In this article, we study the mural restoration work based on artificial intelligence-assisted multiscale trace generation. Firstly, we convert the fresco images to colour space to obtain the luminance and chromaticity component images; then we process each component image to enhance the edges of the exfoliated region using high and low hat operations; then we construct a multistructure morphological filter to smooth the noise of the image. Finally, the fused mask image is fused with the original mural to obtain the (...) final calibration result. The fresco is converted to HSV colour space, and chromaticity, saturation, and luminance features are introduced; then the confidence term and data term are used to determine the priority of shedding boundary points; then a new block matching criterion is defined, and the best matching block is obtained to replace the block to be repaired based on the structural similarity between the block to be repaired and the matching block by global search; finally, the restoration result is converted to RGB colour space to obtain the final restoration result. An improved generative adversarial network structure is proposed to address the shortcomings of the existing network structure in mural defect restoration, and the effectiveness of the improved modules of the network is verified. Compared with the existing mural restoration algorithms on the test data experimentally verified, the peak signal-to-noise ratio score is improved by 4% and the structural similarity score is improved by 2%. (shrink)

Sociability is essential for modern robots to increase their acceptability in human environments. Traditional techniques use manually engineered utility functions inspired by observing pedestrian behaviors to achieve social navigation. However, social aspects of navigation are diverse, changing across different types of environments, societies, and population densities, making it unrealistic to use hand-crafted techniques in each domain. This paper presents a data-driven navigation architecture that uses state-of-the-art neural architectures, namely Conditional Neural Processes, to learn global and local controllers of the mobile (...) robot from observations. Additionally, we leverage a state-of-the-art, deep prediction mechanism to detect situations not similar to the trained ones, where reactive controllers step in to ensure safe navigation. Our results demonstrate that the proposed framework can successfully carry out navigation tasks regarding social norms in the data. Further, we showed that our system produces fewer personal-zone violations, causing less discomfort. (shrink)

Este artículo es una revisión de la capacidad de los herbarios fotográficos para establecer posibles alianzas experimentales con potencial para ayudar en la concienciación y resolución de la crisis de biodiversidad vegetal. Se analiza cómo el medio fotográfico, bajo el prisma de la creación artística, puede erigirse como un sistema revelador, capaz de superar la mera descripción y ampliar las limitaciones cognitivas de nuestra percepción visual, desvelando la complejidad del universo botánico mediante una mirada más profunda y poética de su (...) naturaleza física. (shrink)

In this paper, we study the estimation of motion direction prediction for fast motion and propose a threshold-based human target detection algorithm using motion vectors and other data as human target feature information. The motion vectors are partitioned into regions by normalization to form a motion vector field, which is then preprocessed, and then the human body target is detected through its motion vector region block-temporal correlation to detect the human body motion target. The experimental results show that the algorithm (...) is effective in detecting human motion targets in videos with the camera relatively stationary. The algorithm predicts the human body position in the reference frame of the current frame in the video by forward mapping the motion vector of the current frame, then uses the motion vector direction angle histogram as a matching feature, and combines it with a region matching strategy to track the human body target in the predicted region, thus realizing the human body target tracking effect. The algorithm is experimentally proven to effectively track human motion targets in videos with relatively static backgrounds. To address the problem of sample diversity and lack of quantity in a multitarget tracking environment, a generative model based on the conditional variational self-encoder conditional generation of adversarial networks is proposed, and the performance of the generative model is verified using pedestrian reidentification and other datasets, and the experimental results show that the method can take advantage of the advantages of both models to improve the quality of the generated results. (shrink)

This paper aims to highlight the life of computer technologies to understand what kind of ‘technological intentionality’ is present in computers based upon the phenomenological elements constituting the objects in general. Such a study can better explain the effects of new digital technologies on our society and highlight the role of digital technologies by focusing on their activities. Even if Husserlian phenomenology rarely talks about technologies, some of its aspects can be used to address the actions performed by the digital (...) technologies by focusing on the objects’ inner ‘life’ thanks to the analysis of passive synthesis and phenomenological horizons in the objects. These elements can be used in computer technologies to show how digital objects are ‘alive.’ This paper focuses on programs developed through high-order languages like C++ and unsupervised learning techniques like ‘Generative Adversarial Model.’ The phenomenological analysis reveals the computer’s autonomy within the programming stages. At the same time, the conceptual inquiry into the digital system’s learning ability shows the alive and changeable nature of the technological object itself. (shrink)

Learning the structure of event sequences is a ubiquitous problem in cognition and particularly in language. One possible solution is to learn a probabilistic generative model of sequences that allows making predictions about upcoming events. Though appealing from a neurobiological standpoint, this approach is typically not pursued in connectionist modeling. Here, we investigated a sequential version of the restricted Boltzmann machine, a stochastic recurrent neural network that extracts high-order structure from sensory data through unsupervised generative learning and can (...) encode contextual information in the form of internal, distributed representations. We assessed whether this type of network can extract the orthographic structure of English monosyllables by learning a generative model of the letter sequences forming a word training corpus. We show that the network learned an accurate probabilistic model of English graphotactics, which can be used to make predictions about the letter following a given context as well as to autonomously generate high-quality pseudowords. The model was compared to an extended version of simple recurrent networks, augmented with a stochastic process that allows autonomous generation of sequences, and to non-connectionist probabilistic models. We conclude that sequential RBMs and stochastic simple recurrent networks are promising candidates for modeling cognition in the temporal domain. (shrink)

Zero-shot learning is dedicated to solving the classification problem of unseen categories, while generalized zero-shot learning aims to classify the samples selected from both seen classes and unseen classes, in which “seen” and “unseen” classes indicate whether they can be used in the training process, and if so, they indicate seen classes, and vice versa. Nowadays, with the promotion of deep learning technology, the performance of zero-shot learning has been greatly improved. Generalized zero-shot learning is a challenging topic that has (...) promising prospects in many realistic scenarios. Although the zero-shot learning task has made gratifying progress, there is still a strong deviation between seen classes and unseen classes in the existing methods. Recent methods focus on learning a unified semantic-aligned visual representation to transfer knowledge between two domains, while ignoring the intrinsic characteristics of visual features which are discriminative enough to be classified by itself. To solve the above problems, we propose a novel model that uses the discriminative information of visual features to optimize the generative module, in which the generative module is a dual generation network framework composed of conditional VAE and improved WGAN. Specifically, the model uses the discrimination information of visual features, according to the relevant semantic embedding, synthesizes the visual features of unseen categories by using the learned generator, and then trains the final softmax classifier by using the generated visual features, thus realizing the recognition of unseen categories. In addition, this paper also analyzes the effect of the additional classifiers with different structures on the transmission of discriminative information. We have conducted a lot of experiments on six commonly used benchmark datasets. The experimental results show that our model outperforms several state-of-the-art methods for both traditional as well as generalized zero-shot learning. (shrink)

The purpose of crowd counting is to estimate the number of pedestrians in crowd images. Crowd counting or density estimation is an extremely challenging task in computer vision, due to large scale variations and dense scene. Current methods solve these issues by compounding multi-scale Convolutional Neural Network with different receptive fields. In this paper, a novel end-to-end architecture based on Multi-Scale Adversarial Convolutional Neural Network (MSA-CNN) is proposed to generate crowd density and estimate the amount of crowd. Firstly, a (...) multi-scale network is used to extract the globally relevant features in the crowd image, and then fractionally-strided convolutional layers are designed for up-sampling the output to recover the loss of crucial details caused by the earlier max pooling layers. An adversarial loss is directly employed to shrink the estimated value into the realistic subspace to reduce the blurring effect of density estimation. Joint training is performed in an end-to-end fashion using a combination of Adversarial loss and Euclidean loss. The two losses are integrated via a joint training scheme to improve density estimation performance.We conduct some extensive experiments on available datasets to show the significant improvements and supremacy of the proposed approach over the available state-of-the-art approaches. (shrink)

The generative entrenchment of an entity is a measure of how much of the generated structure or activity of a complex system depends upon the presence or activity of that entity. It is argued that entities with higher degrees of generative entrenchment are more conservative in evolutionary changes of such systems. A variety of models of complex structures incorporating the effects of generative entrenchment are presented and we demonstrate their relevance in analyzing and explaining a variety of (...) developmental and evolutionary phenomena, both on a macroscopic developmental and evolutionary scale, and using models and strategies pioneered by Kauffman, on the more microscopic scale appropriate to the analysis of the structure and behavior of gene control networks. The resulting picture suggests that generative entrenchment acts as a powerful and constructive developmental constraint on the course of evolutionary processes. Since virtually any system exhibits varying degrees of generative entrenchment among its parts and activities, these studies and results have in addition broad potential application for the analysis of generative structures in other areas. (shrink)

Discourses on Responsible Innovation and Responsible Research and Innovation, in short RI, have revolved around but not elaborated on the notion of critique. In this article, generative critique is introduced to RI as a practice that sits in-between adversarial armchair critique and co-opted, uncritical service. How to position oneself and be positioned on this spectrum has puzzled humanities scholars and social scientists who engage in interdisciplinary collaborations with scientists, engineers, and other professionals. Recently, generative critique has been (...) presented as a solution to the puzzle in interdisciplinary collaborations on neuroscientific experiments. Generative critique seeks to create connections across disciplines that help remake seemingly stable objects in moments when taken-for-granted ways of seeing and approaching objects are unsettled. In order to translate generative critique from the neurosciences to RI, socio-technical integration research is proposed as a practice of generative critique in interdisciplinary RI collaborations. These collaborations aim to account for societal aspects in research and technology development. For this purpose, a variety of approaches have been developed, including STIR and video-reflexive ethnography. STIR and VRE resemble each other but diverge on affective, collaborative, and temporal dimensions. Their juxtaposition serves to develop suggestions for how STIR could be modified on these dimensions to better enact generative critique in interdisciplinary RI collaborations. In this way, the article contributes to ongoing discussions in RI and in the engaged programme in science and technology studies more broadly on the dynamics of positioning in collaborative work. (shrink)

In this paper, we consider how certain longstanding philosophical questions about mental representation may be answered on the assumption that cognitive and perceptual systems implement hierarchical generative models, such as those discussed within the prediction error minimization framework. We build on existing treatments of representation via structural resemblance, such as those in Gładziejewski :559–582, 2016) and Gładziejewski and Miłkowski, to argue for a representationalist interpretation of the PEM framework. We further motivate the proposed approach to content by arguing that (...) it is consistent with approaches implicit in theories of unsupervised learning in neural networks. In the course of this discussion, we argue that the structural representation proposal, properly understood, has more in common with functional-role than with causal/informational or teleosemantic theories. In the remainder of the paper, we describe the PEM framework for approximate Bayesian inference in some detail, and discuss how structural representations might arise within the proposed Bayesian hierarchies. After explicating the notion of variational inference, we define a subjectively accessible measure of misrepresentation for hierarchical Bayesian networks by appeal to the Kullbach–Leibler divergence between posterior generative and approximate recognition densities, and discuss a related measure of objective misrepresentation in terms of correspondence with the facts. (shrink)

Although discriminative deep neural networks are currently dominant in cognitive modeling, we suggest that capacity-limited, generative models are a promising avenue for future work. Generative models tend to learn both local and global features of stimuli and, when properly constrained, can learn componential representations and response biases found in people's behaviors.

Any approach to the study of managerial situations undertaken without reflection on the underpinning philosophy is flawed because it limits our ability to question the validity of the knowledge claimed in the analysis. The paper considers this issue and presents a philosophical reflection on the use of a systems approach to the modelling of human enterprises. It draws on insights from systems thinking, cognitive science, autopoiesis, communication theory and non-linear dynamics. These are interpreted within the context of social systems as (...) networks of conversations that are generated in language. It is written to invite an exchange of ideas concerning the role of the generative dialogue. (shrink)

The paper questions the compatibility of critical realism with ecological economics. In particular, it is argued that there is radical dissonance between ontological presuppositions of ecological economics and critical realist perspective. The dissonance lies in the need of ecological economics to state strict causal regularities in socio-economic realm, given the environmental intuitions about the nature of economy and the role of materiality and non-human agency in persistence of economic systems. Using conceptual apparatus derived from Andrew Brown’s critique of critical realism (...) and Bruno Latour’s actor-network theory, the paper refuses ontological nature/society dualism employed by critical realism, and stresses the role of non-humans in practical production and reproduction of socio-economic networks on the one hand, and in broadly defined ecological economic research on the other hand. (shrink)

With the advent of Web 2.0, numerous social software applications allow people to publish and share information on the Internet. Two of these types of applications – collaborative workspaces and social network sites – have a number of features in common, which are explored to provide a basis for comparative analysis. This basis is extended with a suitable definition of privacy, a sociological perspective and an applicable adversary model in order to facilitate an investigation of similarities and differences with regard (...) to privacy threats. Practical examples are derived from the use of Wikipedia and Facebook. Analysis suggests that a combination of technical, legal, and normative solutions should be considered to counter privacy issues. A number of potential solutions that may mitigate these issues are proposed. (shrink)

In this work, we investigated a covert communication method in wireless networks, which is realized by multiantenna full-duplex single relay. In the first stage, the source node sends covert messages to the relay, and the relay uses a single antenna to send interference signals to the adversary node to protect the covert information being transmitted. In the second stage, the relay decodes and forwards the covert information received in the first stage; at the same time, the relay uses zero-forcing (...) beamforming to send interference signals to the warden to ensure covert transmission. The detection error rate, transmission outage probability, maximum effective covert rate, and other performance indicators are derived in two stages, and the total performance of the system is derived and analyzed. Then, the performance indicators are verified and analyzed by simulation. Our analysis shows that the maximum effective covert rate of using the characteristics of multiantenna to interfere with Willie in the second stage is taken as the total covert performance of the system, and the transmission interruption probability is significantly less than that of the first stage, so the corresponding maximum effective concealment efficiency will be greater. (shrink)

This article tacks back towards the idealist side of the argument, in a spirited defence of critical humanism against the radical symmetry of ANT. Vandenberghe argues that the critique of reification and the ethics of emancipation require us to go beyond the `flat ontology' of ANT and its intermediate level of sociotechnical networks towards a more stratified view of social reality, which is able to account for the determining effect of broader generative but invisible structures of domination. Reasserting (...) the categorical distinction between the ontological regions inhabited by humans and nonhumans, he develops a critical opposition between the gift economy, which emphasizes qualitative relations of reciprocity between humans and which tends towards the personalization of things, and the commodity economy, which objectifies things as property, promotes the reification of persons, and turns them into strategically operating `humants'. This model is critically applied to ANT by suggesting that its `fetishist' attribution of social power to nonhumans effectively results from a failure to account for the emergent properties of the broader relational and cultural systems in which they are embedded, and which overdetermine the blackboxed object worlds which ANT has described. (shrink)

Tackling grand challenges requires coordination and sustained effort among multiple organizations and stakeholders. Yet research on stakeholder theory has been conceptually constrained in capturing this complexity: existing accounts tend to focus either on dyadic level firm–stakeholder ties or on stakeholder networks within which the focal organization is embedded. We suggest that addressing grand challenges requires a more generative conceptualization of organizations and their constituents as stakeholder systems. Using the metaphor of ballet and insights from dance theory, we highlight (...) four defining dimensions of stakeholder systems (two structural and two dyadic); we proceed to offer a dynamic model of how those dimensions may interact and coevolve. Our metaphor and resulting theory of stakeholder systems are thereby well equipped to incorporate the complexity of tackling grand challenges, where many contemporary stakeholder arrangements are oriented around issues rather than firms. (shrink)

With complex buyer-driven global production networks and a labour-intensive manufacturing process, the fashion industry has become a focal point for debates on the social responsibility of business. Utilising an interview methodology with influential actors from seven export garment manufacturers in Sri Lanka, we explore the situated knowledge at one nodal point of the production network. We conceptualise factory management perspectives on the implementation of corporate social responsibility in terms of the strategic balancing of ethical considerations against the commercial pressures (...) of cost and lead time. Factory managers framed CSR in terms of compliance, rather than going above and beyond regulatory requirements; seeing it as a strategic competitive imperative and less a developmental mechanism. Sri Lankan manufacturers maintain that they have benefitted from a unique combination of factors, including strict national labour laws, an educated workforce, the characteristics of the garments produced, industrial upgrading, and long-term non-adversarial buyer–supplier relationships, which they argue has supported the establishment and maintenance of CSR practices. The paper thus provides managerial implications that relate CSR activities to CSR outcomes which include both reputational and production benefits. Such insights will be of strategic relevance for lead retail buyers as well as apparel producers keen to invest in CSR to partly mitigate against increasing price-based competition. (shrink)

Scholarship flourishes in inclusive environments where open deliberations and generative feedback expand both individual and collective thinking. Many researchers, however, have limited access to such settings, and most conventional academic conferences fall short of promises to provide them. We have written this Field Report to share our methods for cultivating a vibrant intellectual community within the Science and Technology Studies Food and Agriculture Network (STSFAN). This is paired with insights from 21 network members on aspects that have allowed STSFAN (...) to thrive, even amid a global pandemic. Our hope is that these insights will encourage others to cultivate their own intellectual communities, where they too can receive the support they need to deepen their scholarship and strengthen their intellectual relationships. (shrink)

Emotion recognition plays an important part in human-computer interaction. Currently, the main challenge in electroencephalogram -based emotion recognition is the non-stationarity of EEG signals, which causes performance of the trained model decreasing over time. In this paper, we propose a two-level domain adaptation neural network to construct a transfer model for EEG-based emotion recognition. Specifically, deep features from the topological graph, which preserve topological information from EEG signals, are extracted using a deep neural network. These features are then passed through (...) TDANN for two-level domain confusion. The first level uses the maximum mean discrepancy to reduce the distribution discrepancy of deep features between source domain and target domain, and the second uses the domain adversarial neural network to force the deep features closer to their corresponding class centers. We evaluated the domain-transfer performance of the model on both our self-built data set and the public data set SEED. In the cross-day transfer experiment, the ability to accurately discriminate joy from other emotions was high: sadness, anger, and fear on the self-built data set. The accuracy reached 74.93% on the SEED data set. In the cross-subject transfer experiment, the ability to accurately discriminate joy from other emotions was equally high: sadness, anger, and fear on the self-built data set. The average accuracy reached 87.9% on the SEED data set, which was higher than WGAN-DA. The experimental results demonstrate that the proposed TDANN can effectively handle the domain transfer problem in EEG-based emotion recognition. (shrink)