Methods of machine learning (ML) are gradually complementing and sometimes even replacing methods of classical statistics in science. This raises the question whether ML faces the same methodological problems as classical statistics. This paper sheds light on this question by investigating a long-standing challenge to classical statistics: the reference class problem (RCP). It arises whenever statistical evidence is applied to an individual object, since the individual belongs to several reference classes and evidence might vary across them. Thus, the problem consists (...) in choosing a suitable reference class for the individual. I argue that deep neural networks (DNNs) are able to overcome specific instantiations of the RCP. Whereas the criteria of narrowness, reliability, and homogeneity, that have been proposed to determine a suitable reference class, pose an inextricable tradeoff to classical statistics, DNNs are able to satisfy them in some situations. On the one hand, they can exploit the high dimensionality in big-data settings. I argue that this corresponds to the criteria of narrowness and reliability. On the other hand, ML research indicates that DNNs are generally not susceptible to overfitting. I argue that this property is related to a particular form of homogeneity. Taking both aspects together reveals that there are specific settings in which DNNs can overcome the RCP. (shrink)

An ideal vision model accounts for behavior and neurophysiology in both naturalistic conditions and designed lab experiments. Unlike psychological theories, artificial neural networks (ANNs) actually perform visual tasks and generate testable predictions for arbitrary inputs. These advantages enable ANNs to engage the entire spectrum of the evidence. Failures of particular models drive progress in a vibrant ANN research program of human vision.

Legal text retrieval serves as a key component in a wide range of legal text processing tasks such as legal question answering, legal case entailment, and statute law retrieval. The performance of legal text retrieval depends, to a large extent, on the representation of text, both query and legal documents. Based on good representations, a legal text retrieval model can effectively match the query to its relevant documents. Because legal documents often contain long articles and only some parts are relevant (...) to queries, it is quite a challenge for existing models to represent such documents. In this paper, we study the use of attentive neural network-based text representation for statute law document retrieval. We propose a general approach using deep neural networks with attention mechanisms. Based on it, we develop two hierarchical architectures with sparse attention to represent long sentences and articles, and we name them Attentive CNN and Paraformer. The methods are evaluated on datasets of different sizes and characteristics in English, Japanese, and Vietnamese. Experimental results show that: (i) Attentive neural methods substantially outperform non-neural methods in terms of retrieval performance across datasets and languages; (ii) Pretrained transformer-based models achieve better accuracy on small datasets at the cost of high computational complexity while lighter weight Attentive CNN achieves better accuracy on large datasets; and (iii) Our proposed Paraformer outperforms state-of-the-art methods on COLIEE dataset, achieving the highest recall and F2 scores in the top-N retrieval task. (shrink)

Although the management and planning of water resources are extremely significant to human development, the complexity of implementation is unimaginable. To achieve this, the high-precision water consumption prediction is actually the key component of urban water optimization management system. Water consumption is usually affected by many factors, such as weather, economy, and water prices. If these impact factors are directly combined to predict water consumption, the weight of each perspective on the water consumption will be ignored, which will be greatly (...) detrimental to the prediction accuracy. Therefore, this paper proposes a deep neural network-based complex system for urban water management. The essence of it is to formulate a water consumption prediction model with the aid of principal component analysis and the integrated deep neural network, which is abbreviated as UWM-Id. The PCA classifies the factors affecting water consumption in the original data into three categories according to their correlation and inputs them into the neural network model. The results in the previous step are assigned weights and integrated into the form of fully connected layer. Finally, analyzing the sensitivity of the proposed UWM-Id and comparing its performance with a series of commonly used baseline methods for data mining, a large number of experiments have proved that UWM-Id has good performance and can be used for urban water management system. (shrink)

The number of comments/reviews for movies is enormous and cannot be processed manually. Therefore, machine learning techniques are used to efficiently process the user’s opinion. This research work proposes a deep neural network with seven layers for movie reviews’ sentiment analysis. The model consists of an input layer called the embedding layer, which represents the dataset as a sequence of numbers called vectors, and two consecutive layers of 1D-CNN for extracting features. A global max-pooling layer is used to (...) reduce dimensions. A dense layer for classification and a dropout layer are also used to reduce overfitting and improve generalization error in the neural network. A fully connected layer is the last layer to predict between two classes. Two movie review datasets are used and widely accepted by the research community. The first dataset contains 25,000 samples, half positive and half negative, whereas the second dataset contains 50,000 specimens of movie reviews. Our neural network model performs sentiment classification among positive and negative movie reviews called binary classification. The model achieves 92% accuracy on both datasets, which is more efficient than traditional machine learning models. (shrink)

Deep neural networks are said to be opaque, impeding the development of safe and trustworthy artificial intelligence, but where this opacity stems from is less clear. What are the sufficient properties for neural network opacity? Here, I discuss five common properties of deep neural networks and two different kinds of opacity. Which of these properties are sufficient for what type of opacity? I show how each kind of opacity stems from only one of (...) these five properties, and then discuss to what extent the two kinds of opacity can be mitigated by explainability methods. (shrink)

Emotion-based sentimental analysis has recently received a lot of interest, with an emphasis on automated identification of user behavior, such as emotional expressions, based on online social media texts. However, the majority of the prior attempts are based on traditional procedures that are insufficient to provide promising outcomes. In this study, we categorize emotional sentiments by recognizing them in the text. For that purpose, we present a deep learning model, bidirectional long-term short-term memory, for emotion recognition that takes into (...) account five main emotions. We use our experimental assessments on the emotion dataset to accomplish the emotion categorization job. The datasets were evaluated and the findings revealed that, when compared to state-of-the-art methodologies, the proposed model can successfully categorize user emotions into several classifications. Finally, we assess the efficacy of our strategy using statistical analysis. This research’s findings help firms to apply best practices in the selection, management, and optimization of policies, services, and product information. (shrink)

The ability to conduct logical reasoning is a fundamental aspect of intelligent behavior, and thus an important problem along the way to human-level artificial intelligence. Traditionally, symbolic methods from the field of knowledge representation and reasoning have been used to equip agents with capabilities that resemble human reasoning qualities. More recently, however, there has been an increasing interest in applying alternative approaches based on machine learning rather than logic-based formalisms to tackle this kind of tasks. Here, we make use of (...) state-of-the-art methods for training deep neural networks to devise a novel model that is closely coupled to symbolic reasoning methods, and thus able to learn how to effectively perform basic ontology reasoning. This term describes an important and at the same time very natural kind of problem settings where the rules for conducting reasoning are specified alongside with the actual information. Many problems in practice may be viewed as such reasoning tasks, which is why the presented approach is applicable to a plethora of important real-world problems. To demonstrate the effectiveness of the suggested method, we present the outcomes of several experiments that have been conducted on both toy datasets as well as real-world data, which show that our model learned to perform precise reasoning on a number of diverse inference tasks that require comprehensive deductive proficiencies. Furthermore, it turned out that the suggested model suffers much less from different obstacles that prohibit symbolic reasoning. (shrink)

The ability to conduct logical reasoning is a fundamental aspect of intelligent behavior, and thus an important problem along the way to human-level artificial intelligence. Traditionally, symbolic methods from the field of knowledge representation and reasoning have been used to equip agents with capabilities that resemble human reasoning qualities. More recently, however, there has been an increasing interest in applying alternative approaches based on machine learning rather than logic-based formalisms to tackle this kind of tasks. Here, we make use of (...) state-of-the-art methods for training deep neural networks to devise a novel model that is closely coupled to symbolic reasoning methods, and thus able to learn how to effectively perform basic ontology reasoning. This term describes an important and at the same time very natural kind of problem settings where the rules for conducting reasoning are specified alongside with the actual information. Many problems in practice may be viewed as such reasoning tasks, which is why the presented approach is applicable to a plethora of important real-world problems. To demonstrate the effectiveness of the suggested method, we present the outcomes of several experiments that have been conducted on both toy datasets as well as real-world data, which show that our model learned to perform precise reasoning on a number of diverse inference tasks that require comprehensive deductive proficiencies. Furthermore, it turned out that the suggested model suffers much less from different obstacles that prohibit symbolic reasoning. (shrink)

Depending on what we mean by “explanation,” challenges to the explanatory depth and reach of deep neural network models of visual and other forms of intelligent behavior may need revisions to both the elementary building blocks of neural nets (the explananda) and to the ways in which experimental environments and training protocols are engineered (the explanantia). The two paths assume and imply sharply different conceptions of how an explanation explains and of the explanatory function of models.

Bowers et al. argue that deep neural networks (DNNs) are poor models of biological vision because they often learn to rival human accuracy by relying on strategies that differ markedly from those of humans. We show that this problem is worsening as DNNs are becoming larger-scale and increasingly more accurate, and prescribe methods for building DNNs that can reliably model biological vision.

“Rationality” in Simon's “bounded rationality” is the principle that humans make decisions on the basis of step-by-step reasoning using systematic rules of logic to maximize utility. “Bounded rationality” is the observation that the ability of a human brain to handle algorithmic complexity and large quantities of data is limited. Bounded rationality, in other words, treats a decision maker as a machine carrying out computations with limited resources. Under the principle of embodied cognition, a cognitive mind is an interactive machine. Turing-Church (...) computations are not interactive, and interactive machines can accomplish things that no Turing-Church computation can accomplish. Hence, if “rationality” is computation, and “bounded rationality” is computation with limited complexity, then “embodied bounded rationality” is both more limited than computation and more powerful. By embracing interaction, embodied bounded rationality can accomplish things that Turing-Church computation alone cannot. Deep neural networks, which have led to a revolution in artificial intelligence, are both interactive and not fundamentally algorithmic. Hence, their ability to mimic some cognitive capabilities far better than prior algorithmic techniques based on symbol manipulation provides empirical evidence for the principle of embodied bounded rationality. (shrink)

Decades of psychological research have been aimed at modeling how people learn features and categories. The empirical validation of these theories is often based on artificial stimuli with simple representations. Recently, deep neural networks have reached or surpassed human accuracy on tasks such as identifying objects in natural images. These networks learn representations of real‐world stimuli that can potentially be leveraged to capture psychological representations. We find that state‐of‐the‐art object classification networks provide surprisingly accurate predictions (...) of human similarity judgments for natural images, but they fail to capture some of the structure represented by people. We show that a simple transformation that corrects these discrepancies can be obtained through convex optimization. We use the resulting representations to predict the difficulty of learning novel categories of natural images. Our results extend the scope of psychological experiments and computational modeling by enabling tractable use of large natural stimulus sets. (shrink)

The predominance of machine learning based techniques in cognitive neuroscience raises a host of philosophical and methodological concerns. Given the messiness of neural activity, modellers must make choices about how to structure their raw data to make inferences about encoded representations. This leads to a set of standard methodological assumptions about when abstraction is appropriate in neuroscientific practice. Yet, when made uncritically these choices threaten to bias conclusions about phenomena drawn from data. Contact between the practices of multivariate pattern (...) analysis and philosophy of science can help to illuminate the conditions under which we can use artificial neural networks to better understand neural mechanisms. This paper considers a specific technique for MVPA called representational similarity analysis. I develop a theoretically-informed account of RSA that draws on early connectionist research and work on idealization in the philosophy of science. By bringing a philosophical account of cognitive modelling in conversation with RSA, this paper clarifies the practices of neuroscientists and provides a generalizable framework for using artificial neural networks to study neural mechanisms in the brain. (shrink)

Deep neural networks (DNNs) have had extraordinary successes in classifying photographic images of objects and are often described as the best models of biological vision. This conclusion is largely based on three sets of findings: (1) DNNs are more accurate than any other model in classifying images taken from various datasets, (2) DNNs do the best job in predicting the pattern of human errors in classifying objects taken from various behavioral datasets, and (3) DNNs do the best (...) job in predicting brain signals in response to images taken from various brain datasets (e.g., single cell responses or fMRI data). However, these behavioral and brain datasets do not test hypotheses regarding what features are contributing to good predictions and we show that the predictions may be mediated by DNNs that share little overlap with biological vision. More problematically, we show that DNNs account for almost no results from psychological research. This contradicts the common claim that DNNs are good, let alone the best, models of human object recognition. We argue that theorists interested in developing biologically plausible models of human vision need to direct their attention to explaining psychological findings. More generally, theorists need to build models that explain the results of experiments that manipulate independent variables designed to test hypotheses rather than compete on making the best predictions. We conclude by briefly summarizing various promising modeling approaches that focus on psychological data. (shrink)

Face recognition is a computationally challenging classification task. Deep convolutional neural networks (DCNNs) are brain‐inspired algorithms that have recently reached human‐level performance in face and object recognition. However, it is not clear to what extent DCNNs generate a human‐like representation of face identity. We have recently revealed a subset of facial features that are used by humans for face recognition. This enables us now to ask whether DCNNs rely on the same facial information and whether this human‐like (...) representation depends on a system that is optimized for face identification. In the current study, we examined the representation of DCNNs of faces that differ in features that are critical or non‐critical for human face recognition. Our findings show that DCNNs optimized for face identification are tuned to the same facial features used by humans for face recognition. Sensitivity to these features was highly correlated with performance of the DCNN on a benchmark face recognition task. Moreover, sensitivity to these features and a view‐invariant face representation emerged at higher layers of a DCNN optimized for face recognition but not for object recognition. This finding parallels the division to a face and an object system in high‐level visual cortex. Taken together, these findings validate human perceptual models of face recognition, enable us to use DCNNs to test predictions about human face and object recognition as well as contribute to the interpretability of DCNNs. (shrink)

Deep Neural Networks (DNNs) are becoming increasingly important as scientific tools, as they excel in various scientific applications beyond what was considered possible. Yet from a certain vantage point, they are nothing but parametrized functions fθ(x)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{f}_{\varvec{\theta }}(\varvec{x})$$\end{document} of some data vector x\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{x}$$\end{document}, and their ‘learning’ is nothing but an iterative, algorithmic fitting of the parameters to data. Hence, what could (...) be special about them as a scientific tool or model? I will here suggest an integrated perspective that mediates between extremes, by arguing that what makes DNNs in science special is their ability to develop functional concept proxies (FCPs): Substructures that occasionally provide them with abilities that correspond to those facilitated by concepts in human reasoning. Furthermore, I will argue that this introduces a problem that has so far barely been recognized by practitioners and philosophers alike: That DNNs may succeed on some vast and unwieldy data sets because they develop FCPs for features that are not transparent to human researchers. The resulting breach between scientific success and human understanding I call the ‘Actually Smart Hans Problem’. (shrink)

Cardiac disease is the leading cause of death worldwide. Cardiovascular diseases can be prevented if an effective diagnostic is made at the initial stages. The ECG test is referred to as the diagnostic assistant tool for screening of cardiac disorder. The research purposes of a cardiac disorder detection system from 12-lead-based ECG Images. The healthcare institutes used various ECG equipment that present results in nonuniform formats of ECG images. The research study proposes a generalized methodology to process all formats of (...) ECG. Single Shoot Detection MobileNet v2-based Deep Neural Network architecture was used to detect cardiovascular disease detection. The study focused on detecting the four major cardiac abnormalities with 98% accuracy results were calculated. The work is relatively rare based on their dataset; a collection of 11,148 standard 12-lead-based ECG images used in this study were manually collected from health care institutes and annotated by the domain experts. The study achieved high accuracy results to differentiate and detect four major cardiac abnormalities. Several cardiologists manually verified the proposed system’s accuracy result and recommended that the proposed system can be used to screen for a cardiac disorder. (shrink)

Theoretical research into the emotional attributes of ideological and political education can improve our ability to understand human emotion and solve socio-emotional problems. To that end, this study undertook an analysis of emotion in ideological and political education by integrating a gate recurrent unit with an attention mechanism. Based on the good results achieved by BERT in the downstream network, we use the long focusing attention mechanism assisted by two-way GRU to extract relevant information and global information of ideological and (...) political education and emotion analysis, respectively. The two kinds of information complement each other, and the accuracy of emotion information can be further improved by combining neural network model. Finally, the validity and domain adaptability of the model were verified using several publicly available, fine-grained emotion datasets. (shrink)

Psychologically faithful deep neural networks (DNNs) could be constructed by training with psychophysics data. Moreover, conventional DNNs are mostly monocular vision based, whereas the human brain relies mainly on binocular vision. DNNs developed as smaller vision agent networks associated with fundamental and less intelligent visual activities, can be combined to simulate more intelligent visual activities done by the biological brain.

Machine learning systems are making considerable inroads in society owing to their ability to recognize and predict patterns. However, the decision-making logic of some widely used machine learning models, such as deep neural networks, is characterized by opacity, thereby rendering them exceedingly difficult for humans to understand and explain and, as a result, potentially risky to use. Considering the importance of addressing this opacity, this paper calls for research that studies empirically and theoretically how machine learning experts (...) and users seek to attain machine learning explainability. Focusing on automated trading, we take steps in this direction by analyzing a trading firm’s quest for explaining its deep neural network system’s actionable predictions. We demonstrate that this explainability effort involves a particular form of human–machine interaction that contains both anthropomorphic and technomorphic elements. We discuss this attempt to attain machine learning explainability in light of reflections on cross-species companionship and consider it an example of human–machine companionship. (shrink)

In today’s era of rapid development in science and technology, the development of digital technology has increasingly higher requirements for data processing functions. The matrix signal commonly used in engineering applications also puts forward higher requirements for processing speed. The eigenvalues of the matrix represent many characteristics of the matrix. Its mathematical meaning represents the expansion of the inherent vector, and its physical meaning represents the spectrum of vibration. The eigenvalue of a matrix is the focus of matrix theory. The (...) problem of matrix eigenvalues is widely used in many research fields such as physics, chemistry, and biology. A neural network is a neuron model constructed by imitating biological neural networks. Since it was proposed, the application research of its typical models, such as recurrent neural networks and cellular neural networks, has become a new hot spot. With the emergence of deep neural network theory, scholars continue to combine deep neural networks to calculate matrix eigenvalues. This article aims to study the estimation and application of matrix eigenvalues based on deep neural networks. This article introduces the related methods of matrix eigenvalue estimation based on deep neural networks, and also designs experiments to compare the time of matrix eigenvalue estimation methods based on deep neural networks and traditional algorithms. It was found that under the serial algorithm, the algorithm based on the deep neural network reduced the calculation time by about 7% compared with the traditional algorithm, and under the parallel algorithm, the calculation time was reduced by about 17%. Experiments are also designed to calculate matrix eigenvalues with Obj and recurrent neural networks (RNNS) models, which proves that the Oja algorithm is only suitable for calculating the maximum eigenvalues of non-negative matrices, while RNNS is commonly used in general models. (shrink)

Internet-of-Things (IoT) creates a significant impact in spectrum sensing, information retrieval, medical analysis, traffic management, etc. These applications require continuous information to perform a specific task. At the time, various intermediate attacks such as jamming, priority violation attacks, and spectrum poisoning attacks affect communication because of the open nature of wireless communication. These attacks create security and privacy issues while making data communication. Therefore, a new method autoencoder deep neural network (AENN) is developed by considering exploratory, evasion, causative, (...) and priority violation attack. The created method classifies the transmission outcomes used to predict the transmission state, whether it is jam data transmission or sensing data. After that, the sensing data is applied for network training that predicts the intermediate attacks. In addition to this, the channel access algorithm is used to validate the channel for every access that minimizes unauthorized access. After validating the channel according to the neural network, data have been transmitted over the network. The defined process is implemented, and the system minimizes different attacks on various levels of energy consumption. The effectiveness of the system is implemented using TensorFlow, and the system ensures the 99.02% of detection rate when compared with other techniques. (shrink)

In this study, the focus was on the development of green energy and future prediction for the consumption of current energy sources and green energy development using an improved deep learning algorithm. In addition to the analysis of the current energy consumption used for the natural gas and oil as fuel, deep neural network algorithm is used to train the system as well as to process the data obtained previously, ranging from literature from the year 2003 until (...) the year 2019, for consumption of fuel. Also, using the proposed algorithm to predict the development of green energy consumption till 2030 is presented in terms of solar and wind generators. The resulting study also focuses on depletion of energy currently used or pollution caused because of it. The green energy controlling issue can take effect by using multiple layers of handling different features extracted from different sources and then learning the system to control it.This study aims to take advantage of carbon emissions to reduce their impact and dependence in the future on environmentally friendly renewable energies. Predicting the correct and precise amount of energy consumption and increasing the amount of environmentally friendly energy lead to a healthy ecosystem. The expected green energy consumption in the future is almost 78.25 EJ in 2030 and will be, in total energy average, 56% in 2045. The aim is to reduce dependency on costly and environmentally harmful fuels. (shrink)

In the area of machine learning, different techniques are used to train machines and perform different tasks like computer vision, data analysis, natural language processing, and speech recognition. Computer vision is one of the main branches where machine learning and deep learning techniques are being applied. Optical character recognition is the ability of a machine to recognize the character of a language. Pashto is one of the most ancient and historical languages of the world, spoken in Afghanistan and Pakistan. (...) OCR application has been developed for various cursive languages like Urdu, Chinese, and Japanese, but very little work is done for the recognition of the Pashto language. When it comes to handwritten character recognition, it becomes more difficult for OCR to recognize the characters as every handwritten character’s shape is influenced by the writer’s hand motion dynamics. The reason for the lack of research in Pashto handwritten character data as compared to other languages is because there is no benchmark dataset available for experimental purposes. This study focuses on the creation of such a dataset, and then for the evaluation purpose, a machine is trained to correctly recognize unseen Pashto handwritten characters. To achieve this objective, a dataset of 43000 images was created. Three Feed Forward Neural Network models with backpropagation algorithm using different Rectified Linear Unit layer configurations were trained and tested with this dataset. The simulation shows that Model 1 achieved accuracy up to 87.6% on unseen data while Model 2 achieved an accuracy of 81.60% and 3% accuracy, respectively. Similarly, loss was the lowest for Model 1 with 0.15 and 3.17 for training and testing, followed by Model 2 with 0.7 and 4.2 for training and testing, while Model 3 was the last with loss values of 6.4 and 3.69. The precision, recall, and f-measure values of Model 1 were better than those of both Model 2 and Model 3. Based on results, Model 1 is found to be the most efficient as compared to the other two models in terms of accuracy to recognize Pashto handwritten characters. (shrink)

The annotated dataset is an essential requirement to develop an artificial intelligence system effectively and expect the generalization of the predictive models and to avoid overfitting. Lack of the training data is a big barrier so that AI systems can broaden in several domains which have no or missing training data. Building these datasets is a tedious and expensive task and depends on the domains and languages. This is especially a big challenge for low-resource languages. In this paper, we experiment (...) and evaluate many various approaches on sentiment analysis problems so that they can still obtain high performances under limited training data. This paper uses the preprocessing techniques to clean and normalize the data and generate the new samples from the limited training dataset based on many text augmentation techniques such as lexicon substitution, sentence shuffling, back translation, syntax-tree transformation, and embedding mixup. Several experiments have been performed for both well-known machine learning-based classifiers and deep learning models. We compare, analyze, and evaluate the results to indicate the advantage and disadvantage points of the techniques for each approach. The experimental results show that the data augmentation techniques enhance the accuracy of the predictive models; this promises that smart systems can be applied widely in several domains under limited training data. (shrink)

Since entering the new century, People’s living standards are constantly improving, with the continuous improvement of living conditions, people are becoming more and more important in education, which is the embodiment of the enhancement of national strength. The education level is getting higher and higher, and a good education level needs a good teacher–student relationship. To solve these problems, we use the emotional cognition of God’s network to study the teacher–student relationship, and collect and analyze the data of the teacher–student (...) relationship. In this chapter, we use GABP neural network algorithm DHNN algorithm and discrete Hopfield neural network to make the collected data more convenient to be analyzed. The research shows that there is a close relationship between the educational level and the relationship between teachers and students in China, and a good relationship between teachers and students will promote the improvement of educational level. According to the research data, “face-to-face” is the most important way of interaction between tutors and postgraduates in various types of teacher–student relationship. QQ WeChat is also one of the main ways of interaction between students and teachers, which shows that the interaction between students and teachers is talking about the interaction between online and Internet. The education industry is becoming more and more important, and the teacher–student relationship is the most important part of the education industry. Good teacher–student relationship is helpful to cultivate students’ healthy personality. In view of the cold relationship between teachers and students at present, we need to make some measures the relationship between teachers and students and effectively use the relationship between teachers and students to promote the better development of the education industry. (shrink)

Given the extent of using deep convolutional neural networks to model the mechanism of object recognition, it becomes important to analyse the evidence of their similarity and the explanatory potential of these models. I focus on one frequent method of their comparison—representational similarity analysis, and I argue, first, that it underdetermines these models as how-actually mechanistic explanations. This happens because different similarity measures in this framework pick out different mechanisms across DCNNs and the brain in order to (...) correspond them, and there is no arbitration between them in terms of relevance for object recognition. Second, the reason similarity measures are underdetermining to a large degree stems from the highly idealised nature of these models, which undermines their status as how-possibly mechanistic explanatory models of object recognition as well. Thus, building models with more theoretical consideration and choosing relevant similarity measures may bring us closer to the goal of mechanistic explanation. (shrink)

In artificial intelligence, recent research has demonstrated the remarkable potential of Deep Convolutional Neural Networks (DCNNs), which seem to exceed state-of-the-art performance in new domains weekly, especially on the sorts of very difficult perceptual discrimination tasks that skeptics thought would remain beyond the reach of artificial intelligence. However, it has proven difficult to explain why DCNNs perform so well. In philosophy of mind, empiricists have long suggested that complex cognition is based on information derived from sensory experience, (...) often appealing to a faculty of abstraction. Rationalists have frequently complained, however, that empiricists never adequately explained how this faculty of abstraction actually works. In this paper, I tie these two questions together, to the mutual benefit of both disciplines. I argue that the architectural features that distinguish DCNNs from earlier neural networks allow them to implement a form of hierarchical processing that I call “transformational abstraction”. Transformational abstraction iteratively converts sensory-based representations of category exemplars into new formats that are increasingly tolerant to “nuisance variation” in input. Reflecting upon the way that DCNNs leverage a combination of linear and non-linear processing to efficiently accomplish this feat allows us to understand how the brain is capable of bi-directional travel between exemplars and abstractions, addressing longstanding problems in empiricist philosophy of mind. I end by considering the prospects for future research on DCNNs, arguing that rather than simply implementing 80s connectionism with more brute-force computation, transformational abstraction counts as a qualitatively distinct form of processing ripe with philosophical and psychological significance, because it is significantly better suited to depict the generic mechanism responsible for this important kind of psychological processing in the brain. (shrink)

Deep neural networks (DNNs) have found many useful applications in recent years. Of particular interest have been those instances where their successes imitate human cognition and many consider artificial intelligences to offer a lens for understanding human intelligence. Here, we criticize the underlying conflation between the predictive and explanatory power of DNNs by examining the goals of modeling.

The skin, uniquely positioned at the interface between the human body and the external world, plays a multifaceted immunologic role in human life. In medical practice, early accurate detection of all types of skin tumours is essential to guide appropriate management and improve patients’ survival. The most important issue is to differentiate between malignant skin tumours and benign lesions. The aim of this research is the classification of skin tumours by analysing medical skin tumour dermoscopy images. This paper is focused (...) on a new strategy based on deep convolutional neural networks which have recently shown a state-of-the-art performance to define strategy to automatic classification for skin tumour images. The proposed system is tested on well-known HAM10000 data set. For experimental results, verification is performed and the results are compared with similar researches. (shrink)

Early and accurate diagnosis of glaucoma is critical for avoiding human vision deterioration and preventing blindness. A deep-neural-network model has been developed for the diagnosis of glaucoma based on Heidelberg retina tomography, called “Seeking Common Features and Reserving Differences Net” to make full use of the HRT data. In this work, the proposed SCRD-Net model achieved an area under the curve of 94.0%. For the two HRT image modalities, the model sensitivities were 91.2% and 78.3% at specificities of (...) 0.85 and 0.95, respectively. These results demonstrate a significant improvement over earlier results. In addition, we visualized the network outputs to develop an interpretation of the learned mechanism for discriminating glaucoma and normal images. Thus, the SCRD-Net can be an effective diagnostic indicator of glaucoma during clinical screening. To facilitate SCRD-Net utilization by the scientific community, the code implementation will be made publicly available. (shrink)

This paper proposes several recurrent neural network-based models for recognizing requisite and effectuation parts in Legal Texts. Firstly, we propose a modification of BiLSTM-CRF model that allows the use of external features to improve the performance of deep learning models in case large annotated corpora are not available. However, this model can only recognize RE parts which are not overlapped. Secondly, we propose two approaches for recognizing overlapping RE parts including the cascading approach which uses the sequence of (...) BiLSTM-CRF models and the unified model approach with the multilayer BiLSTM-CRF model and the multilayer BiLSTM-MLP-CRF model. Experimental results on two Japan law RRE datasets demonstrated advantages of our proposed models. For the Japanese National Pension Law dataset, our approaches obtained an \ score of 93.27% and exhibited a significant improvement compared to previous approaches. For the Japan Civil Code RRE dataset which is written in English, our approaches produced an \ score of 78.24% in recognizing RE parts that exhibited a significant improvement over strong baselines. In addition, using external features and in-domain pre-trained word embeddings also improved the performance of RRE systems. (shrink)

Faces provide information about a person's identity, as well as their sex, age, and ethnicity. People also infer social and personality traits from the face — judgments that can have important societal and personal consequences. In recent years, deep convolutional neural networks (DCNNs) have proven adept at representing the identity of a face from images that vary widely in viewpoint, illumination, expression, and appearance. These algorithms are modeled on the primate visual cortex and consist of multiple processing (...) layers of simulated neurons. Here, we examined whether a DCNN trained for face identification also retains a representation of the information in faces that supports social‐trait inferences. Participants rated male and female faces on a diverse set of 18 personality traits. Linear classifiers were trained with cross validation to predict human‐assigned trait ratings from the 512 dimensional representations of faces that emerged at the top‐layer of a DCNN trained for face identification. The network was trained with 494,414 images of 10,575 identities and consisted of seven layers and 19.8 million parameters. The top‐level DCNN features produced by the network predicted the human‐assigned social trait profiles with good accuracy. Human‐assigned ratings for the individual traits were also predicted accurately. We conclude that the face representations that emerge from DCNNs retain facial information that goes beyond the strict limits of their training. (shrink)

Role-based database management systems are most widely used for information storage and analysis but are known as vulnerable to insider attacks. The core of intrusion detection lies in an adaptive system, where an insider attack can be judged if it is different from the predicted role by performing classification on the user’s queries accessing the database and comparing it with the authorized role. In order to handle the high similarity of user queries for misclassified roles, this paper proposes a (...) class='Hi'>deep metric neural network with strategic sampling algorithm that properly extracts salient features and directly learns a quantitative measure of similarity. A strategic sampling method of heuristically generating and learning training pairs through Monte Carlo search is proposed to select a training pair that can represent the entire dataset. With the TPC-E–based benchmark data trained with 11,000 queries for 11 roles, the proposed model produces the classification accuracy of 95.41%, which is the highest compared with the previous models. The results are verified through comparison of quantitative and qualitative evaluations, and the feature space modelled in the neural network is analysed by t-SNE algorithm. (shrink)

The investigation of visual categorization has recently been aided by the introduction of deep convolutional neural networks (CNNs), which achieve unprecedented accuracy in picture classification after extensive training. Even if the architecture of CNNs is inspired by the organization of the visual brain, the similarity between CNN and human visual processing remains unclear. Here, we investigated this issue by engaging humans and CNNs in a two‐class visual categorization task. To this end, pictures containing animals or vehicles were (...) modified to contain only low/high spatial frequency (HSF) information, or were scrambled in the phase of the spatial frequency spectrum. For all types of degradation, accuracy increased as degradation was reduced for both humans and CNNs; however, the thresholds for accurate categorization varied between humans and CNNs. More remarkable differences were observed for HSF information compared to the other two types of degradation, both in terms of overall accuracy and image‐level agreement between humans and CNNs. The difficulty with which the CNNs were shown to categorize high‐passed natural scenes was reduced by picture whitening, a procedure which is inspired by how visual systems process natural images. The results are discussed concerning the adaptation to regularities in the visual environment (scene statistics); if the visual characteristics of the environment are not learned by CNNs, their visual categorization may depend only on a subset of the visual information on which humans rely, for example, on low spatial frequency information. (shrink)