The energy sector contributes approximately two-thirds of global greenhouse gas emissions. In this context, the sector must adapt to new supply and demand networks for all future energy sources. The ongoing transformation in the European energy field is driven by the ambition of the European Union to reach the climate objectives set for 2030. The main actions are increasing renewable energy production, adapting transition fuels like natural gas to reduce emissions, improving energy efficiency across all economic sectors, prioritizing building, transportation, (...) and industry, developing Carbon Capture and Storage technologies, and ensuring universal access to clean and affordable energy. The significant changes envisaged in the energy sector to increase renewable energy production and consumption require improved integration and more use of predictive tools to support stakeholders' decision-making processes. This article presents a case study to assess the performance of predictive models based on a Feed-Forward Neural Network architecture that employ Root Mean Squared Propagation as their optimization function in terms of choosing the most appropriate activation function for this kind of data input and output. The training and testing phases of the models use data about building energy consumption. The lowest training time and Root Mean Square Error values were similar for Rectified Linear Unit and Tanh activation functions. The model with the Rectified linear Unit activation function performed best. (shrink)

Flow prediction in a vegetated channel has been extensively studied in the past few decades. A number of equations that essentially differ from each other in derivation and form have been developed. Because the process is extremely complex, getting the deterministic or analytical form of the process phenomena is too difficult. Hybrid neural network model is particularly useful in modeling processes where an adequate knowledge of the physics is limited. This hybrid model is presented here as a complementary (...) tool to model channel flow–vegetation interactions in submerged vegetation conditions. The hybrid model is used to overcome the local minima limitations of a feed-forward neural network. The prediction capability of model has been found to be better than past empirical predictors. The model developed herein showed significantly better results in several model performance criteria compared with empirical models. (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)

There are many real-world applications like healthcare systems, job monitoring, well-being and personal fitness tracking, monitoring of elderly and frail people, assessment of rehabilitation and follow-up treatments, affording Fall Detection (FD) and ADL (Activity of Daily Living) identification, separately or even at a time. However, the two main drawbacks of these solutions are that most of the times, the devices deployed are obtrusive (devices worn on not quite common parts of the body like neck, waist and ankle) and the poor (...) battery life. Thus, this work proposes a low-power classification algorithm based on an Ensemble of KNN and K-Means algorithms (EKMeans) to identify Falls and High-Intensity ADL events such as running, jogging and climbing up stairs. The input of EKMeans are triaxial accelerometer data gathered from wrist-wearable devices. The proposal will be validated on the Fall&ADL publicly available datasets UMAFall, UCIFall and FallAllD, considering two kinds of activity labelling: Two-Class and Multi-Class. An exhaustive comparative study between our proposal, and the baseline algorithms KNN and a feed-forward Neural Network (NN) is deployed, where EKMeans outperformed clearly the Specificity (ADL classification) of the KNN and NN for the three datasets. Finally, a comparative battery consumption study has been included deploying the analyzed algorithms in a WearOS smartwatch, where EKMeans drains the battery from 100% to 0% in 27.45 hours, saving 5% and 21% concerning KNN and NN, respectively. Keywords: Human Activity Recognition, ADL Identification, Fall Detection TS Clustering, TS Classification, Wearable Devices, Low-Power HAR. (shrink)

Suspended sediment modeling is an important subject for decision-makers at the catchment level. Accurate and reliable modeling of suspended sediment load is important for planning, managing, and designing of water resource structures and river systems. The objective of this study was to develop artificial intelligence- based ensemble methods for modeling SSL in Katar catchment, Ethiopia. In this paper, three single AI-based models, that is, support vector machine, adaptive neurofuzzy inference system, feed-forward neural network, and one conventional (...) multilinear regression modes, were used for SSL modeling. Besides, four different ensemble methods, neural network ensemble, ANFIS ensemble, weighted average ensemble, and simple average ensemble, were developed by combining the outputs of the four single models to improve their predictive performance. The study used two-year discharge and SSL data for training and verification of the applied models. Determination coefficient and root mean square error were used to evaluate the performances of the developed models. Based on the performance measure results, the ANFIS model provides higher efficiency than the other developed single models. Out of all developed ensemble models, the nonlinear ANFIS model combination method was found to be the most accurate method and could increase the efficiency of SVM, MLR, ANFIS, and FFNN models by 19.02%, 37%, 9.73%, and 16.3%, respectively, at the verification stage. Overall, the proposed ensemble models in general and the AI-based ensemble in particular provide excellent performance in SSL estimation. (shrink)

A distinction should be made between the formation of stimulus-driven associations and cognitive concepts. To test the learning mode of a neural network, we propose a simple and classic input-output test: the discrimination shift task. Feed-forward PDP models appear to form stimulus-driven associations. A Hopfield network should be extended to apply the test.

The ability to predict is the most importantability of the brain. Somehow, the cortex isable to extract regularities from theenvironment and use those regularities as abasis for prediction. This is a most remarkableskill, considering that behaviourallysignificant environmental regularities are noteasy to discern: they operate not only betweenpairs of simple environmental conditions, astraditional associationism has assumed, butamong complex functions of conditions that areorders of complexity removed from raw sensoryinputs. We propose that the brain's basicmechanism for discovering such complexregularities is implemented in (...) the dendritictrees of individual pyramidal cells in thecerebral cortex. Pyramidal cells have 5–8principal dendrites, each of which is capableof learning nonlinear input-to-outputtransfer functions. We propose that eachdendrite is trained, in learning its transferfunction, by all the other principal dendritesof the same cell. These dendrites teach eachother to respond to their separate inputs with matching outputs. Exposed to differentbut related information about the sensoryenvironment, principal dendrites of the samecell tune to functions over environmentalconditions that, while different, are correlated . As a result, the cell as awhole tunes to the source of the regularitiesdiscovered by the cooperating dendrites,creating a new representation. When organizedinto feed-forward/feedback layers, pyramidalcells can build their discoveries on thediscoveries of other cells, graduallyuncovering nature's hidden order. Theresulting associative network is powerfulenough to meet a troubling traditionalobjection to associationism: that it is toosimple an architecture to implement rationalprocesses. (shrink)

An incoherent feed‐forward loop (FFL) is one of the most‐frequently observed motifs in biomolecular regulatory networks. It has been thought that the incoherent FFL is designed simply to induce a transient response shaped by a ‘fast activation and delayed inhibition’. We find that the dynamics of various incoherent FFLs can be further classified into two types: time‐dependent biphasic responses and dose‐dependent biphasic responses. Why do the structurally identical incoherent FFLs play such different dynamical roles? Through computational studies, we (...) show that the dynamics of the two types of incoherent FFLs are mutually exclusive. Following from further computational results and experimental observations, we hypothesize that incoherent FFLs have been optimally designed to achieve distinct biological function arising from different cellular contexts. Additional Supporting Information may be found in the online version of the article. BioEssays 30:1204–1211, 2008. © 2008 Wiley Periodicals, Inc. (shrink)

Bowers et al. propose to use controlled behavioral experiments when evaluating deep neural networks as models of biological vision. We agree with the sentiment and draw parallels to the notion that “neuroscience needs behavior.” As a promising path forward, we suggest complementing image recognition tasks with increasingly realistic and well-controlled task environments that engage real-world object recognition behavior.

Yoshimi has attempted to defuse my argument concerning the identification of network abstraction with empiricist abstraction - thus entailing psychologism - by claiming that the argument does not generalize from the example of simple feed-forward networks. I show that the particular details of networks are logically irrelevant to the nature of the abstractive process they employ. This is ultimately because deep artificial neural networks (ANNs) and dynamical systems theory applied to the mind (DST) are both associationisms (...) - that is, empiricist theories that derive the principles of thought from the causal history of the organism/system. On this basis, I put forward a new aspect of the old argument by noting that ANNs & DST are the causal bases of the phenomena of passive synthesis, whereas the language of thought hypothesis (LOT) and the symbolic computational theory of mind (CTM) are the causal bases of the phenomena of active synthesis. If the phenomena of active synthesis are not distinct in kind from and are thus reducible to those of passive synthesis, psychologism results. Yoshimi’s program, insofar as it denies this fundamental phenomenological distinction, is revealed to be the true anti-pluralist program, by essentially denying the causal efficacy of the mechanistic foundations of active synthesis by referring phenomenology exclusively to associationism for its causal foundation. (shrink)

Figure 1: A pr ototyp ical exa mple of a three-layer feed forward network, used by Plunkett and M archm an (1 991 ) to simulate learning the past-tense of En glish verbs. The inpu t units encode representations of the three phonemes of the present tense of the artificial words used in this simulation. Th e netwo rk is trained to produce a representation of the phonemes employed in the past tense form and the suffix (/d/, (...) /ed/, or /t/) used on regular verbs. To run the network, each input unit is assigned an activation value o f 0 or 1 , dep ending on whethe r the featu re is present or not. Eac h input unit is conne cted to each of the 30 hidden units by a we ighted conn ection and p rovid es an inp ut to each hidden unit equal to the product of the input unit's activation and the weight. Each hidd en unit's activation is then determined by summing ov er the va lues co ming fro m each inp ut unit to deter mine a netinput, and then applying a non-linear function (e.g., the logistic function 1/(1+e^netinput). Th is whole proced ure is. (shrink)

Recent experimental findings reveal dissociations of conscious and nonconscious performance in many fields of psychological research, suggesting that conscious and nonconscious effects result from qualitatively different processes. A connectionist view of these processes is put forward in which consciousness is the consequence of construction processes taking place in three types of working memory in a specific type of recurrent neural network. The recurrences arise by feeding back output to the input of a central (representational) network. They (...) are assumed to be intemalizations of motor-sensory feedback through the environment. In this manner, a subvocal-phonological, a visuo-spatial, and a somatosensory working memory may have developed. Representations in the central network, which constitutes long-term memory, can be kept active by rehearsal in the feedback loops. The sequentially recurrent architecture allows for recursive symbolic operations and the formation of (auditory, visual, or somatic) models of the external world which can be maintained, transformed and temporarily combined with other information in working memory. Moreover, the quasi-input from the loop directs subsequent attentional processing. The view may contribute to a formal framework to accommodate findings from disparate fields such as working memory, sequential reasoning, and conscious and nonconscious processes in memory and emotion. In theory, but probably not very soon in practice, such connectionist models might simulate aspects of consciousness. (shrink)

The extended dual-aspect monism framework of consciousness, based on neuroscience, consists of five components: (1) dual-aspect primal entities; (2) neural-Darwinism: co-evolution and co-development of subjective experiences (SEs) and associated neural-nets from the mental aspect (that carries the SEs/proto-experiences (PEs) in superposed and unexpressed form) and the material aspect (mass, charge, spin and space-time) of fundamental entities (elementary particles), respectively and co-tuning via sensorimotor interaction; (3) matching and selection processes: interaction of two modes, namely, (a) the non-tilde mode that (...) is the material and mental aspect of cognition (memory and attention) related feedback signals in a neural-network, which is the cognitive nearest past approaching towards present; and (b) the tilde mode that is the material and mental aspect of the feed forward signals due to external environmental input and internal endogenous input, which is the nearest future approaching towards present and is a entropy-reversed representation of non-tilde mode; (4) the segregation and integration of information, and (5) the necessary ingredients of SEs (such as wakefulness, attention, re-entry, working memory, stimulus at or above threshold level, and neural-net PEs). This framework leads to structural and functional coherence between the mind and the brain, bridges the explanatory gap (the gap between SEs and their neural-correlates), and leads to our mundane subjective experiences. This extended dual-aspect monism (eDAM) framework (Vimal, 2008, 2010, 2013, 2015a, 2015b) could be the fundamental basis of various religions and philosophies. This is a Western perspective. On the other hand, Eastern perspectives emphasize the practical methods for achieving altered experience at samadhi state. An important corollary of these methods (such as yogic method) is the sublimation of negative aspects of seven groups of self-protective energy system (desire, anger, ego, greed, attachment, jealousy, and selfish-love) into their positive aspects. Their negative aspects create war and suffering, whereas their positive aspects advance science and technology, family values, peace, and happiness. Here, the Western perspective framework is extended to include the concepts of the sublimation process to encompass Eastern perspectives. The four elements (war, suffering, peace, and happiness) are ubiquitous in both space and time because they are essential contributors to the variations for natural selection in our evolutionary system. The sublimation process optimizes the system: minimizes war and suffering, maximizes peace and happiness, and enhances family values and individual progress. This is consistent with both Eastern and Western perspectives. (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)

Extreme learning machine is a recent scheme for single hidden layer feed forward networks. It has attracted much interest in the machine intelligence and pattern recognition fields with numerous real-world applications. The ELM structure has several advantages, such as its adaptability to various problems with a rapid learning rate and low computational cost. However, it has shortcomings in the following aspects. First, it suffers from the irrelevant variables in the input data set. Second, choosing the optimal number of (...) neurons in the hidden layer is not well defined. In case the hidden nodes are greater than the training data, the ELM may encounter the singularity problem, and its solution may become unstable. To overcome these limitations, several methods have been proposed within the regularization framework. In this article, we considered a greedy method for sparse approximation of the output weight vector of the ELM network. More specifically, the orthogonal matching pursuit algorithm is embedded to the ELM. This new technique is named OMP-ELM. OMP-ELM has several advantages over regularized ELM methods, such as lower complexity and immunity to the singularity problem. Experimental works on nine commonly used regression problems indicate that the investigated OMP-ELM method confirms these advantages. Moreover, OMP-ELM is compared with the ELM method, the regularized ELM scheme, and artificial neural networks. (shrink)

The application of machine learning touches all activities of human behavior such as computer network and routing packets in LAN. In the field of our research here, emphasis was placed on extracting weights that would affect the speed of the network's response and finding the best path, such as the number of nodes in the path and the congestion on each path, in addition to the cache used for each node. Therefore, the use of these elements in building (...) the neural network is worthy, as is the exploitation of the feed forwarding and the backpropagation in the neural network in order to reach the best prediction for the best path. The goal of the proposed neural network is to minimize the network time delay within the optimization of the packet paths being addressed in this study. The shortest path is considered as the key issue in routing algorithm that can be carried out with real time of path computations. Exploiting the gaps in previous studies, which are represented in the lack of training of the system and the inaccurate prediction as a result of not taking into consideration the hidden layers' feedback, leads to great performance. This study aims to suggest an efficient algorithm that could help in selecting the shortest path to improve the existing methods using weights derived from packet ID and to change neural network iteration simultaneously. In this study, the design of the efficient neural network of appropriate output is discussed in detail including the principles of the network. The findings of the study revealed that exploiting the power of computational system to demonstrate computer simulation is really effective. It is also shown that the system achieved good results when training the neural network system to get 2.4% time delay with 5 nodes in local LAN. Besides, the results showed that the major features of the proposed model will be able to run in real time and are also adaptive to change with path topology. (shrink)

Under the background that the development of low-carbon circular economy is the objective requirement for the in-depth implementation of scientific development and the inevitable choice for promoting the sustainable development of economy and society, it is not only the requirement of corporate social responsibility but also the path to realize corporate social responsibility. Enterprises should become the representative and model of social responsibility practice in the development of low-carbon circular economy, in order to promote the fulfilment and development of corporate (...) social responsibility in the whole society. Therefore, it is of great theoretical and practical significance to study the realization of corporate social responsibility in the context of low-carbon circular economy. This paper introduces the connotation of low-carbon circular economy and corporate social responsibility, analyses the reality and theoretical basis of realizing corporate social responsibility in low-carbon circular economy, analyses the interactive relationship between the development of low-carbon circular economy and the realization of corporate social responsibility, and puts forward the construction of enterprise low-carbon operation mechanism. This paper uses the research of corporate social responsibility based on radial basis function neural network to build a low-carbon circular economy. The evaluation model of environment economy coupling development is verified by an example, which provides useful guidance for the evaluation and development of corporate social responsibility. (shrink)

In the target article, Bowers et al. dispute deep artificial neural network (ANN) models as the currently leading models of human vision without producing alternatives. They eschew the use of public benchmarking platforms to compare vision models with the brain and behavior, and they advocate for a fragmented, phenomenon-specific modeling approach. These are unconstructive to scientific progress. We outline how the Brain-Score community is moving forward to add new model-to-human comparisons to its community-transparent suite of benchmarks.

Differentiating self from other has been investigated at the neural level, and its incorporation into the model proposed Hurley is necessary for the model to be complete. With an emphasis on the feed-forward model in layer 2, we examine the role that self and other disruptions, including auditory verbal hallucinations (AVHs), may have in expanding the model proposed by Hurley.

Recent developments in psychology and neuroscience suggest away to link the mental phenomenon of visual awareness with specific neural processes. Here, it is argued that the feed-forward activation of cells in any area of the brain is not sufficient to generate awareness, but that recurrent processing, mediated by horizontal and feedback connections is necessary. In linking awareness with its neural mechanisms it is furthermore important to dissociate phenomenal awareness from visual attention or decision processes.

This paper deals with the problem of understanding semiosis and meaning in cognitive systems. To this aim we argue for a unified two-factor account according to which both external and internal information are non-independent aspects of meaning, thus contributing as a whole in determining its nature. To overcome the difficulties stemming from this approach we put forward a theoretical scheme based on the definition of a suitable representation space endowed with a set of transformations, and we show how it (...) can be implemented, in the case of a single agent, by a neural network architecture. Numerical experiments conducted on different instances of the latter show that similar representations are developed as a consequence of the fact that these instances are facing a similar semantic task. This allows to model social and environmental influences through a system of interacting agents, each described by a specific implementation of this model architecture. (shrink)

Deep neural network (DNN) models of human-like vision are typically built by feeding blank slate DNN visual images as training data. However, the literature on human perception and perceptual learning suggests that developing DNNs that truly model human vision requires a shift in approach in which perception is not treated as a largely bottom-up process, but as an active, top-down-guided process.

In a matter of mere milliseconds, conversational partners can transform their expectations about the world in a way that accords with another person's perspective. At the same time, in similar situations, the exact opposite also appears to be true. Rather than being at odds, these findings suggest that there are multiple contextual and processing constraints that may guide when and how people consider perspective. These constraints are shaped by a host of factors, including the availability of social and environmental cues, (...) and intrinsic biases and cognitive abilities. To explain how these might be integrated in a new way forward, we turn to an adaptive account of interpersonal interaction. This account draws from basic principles of dynamical systems, principles that we argue are already expressed, both implicitly and explicitly, within a broad landscape of existing research. We then showcase an initial attempt to develop a computational framework to instantiate some of these principles. This framework, consisting of what we argue to be important mechanistic insights rendered by neural network models, is based on a promising and long-standing approach that has yet to take hold in the current domain. We argue that by bridging this gap, new insights into other theoretical accounts, such as the connections between memory and common ground information, might be revealed. (shrink)

With the improvement of citizens’ risk perception ability and environmental protection awareness, social conflicts caused by environmental problems in large-scale construction projects are becoming more and more frequent. Traditional social risk prevention management has some defects in obtaining risk data, such as limited coverage, poor availability, and insufficient timeliness, which makes it impossible to realize effective early warning of social risks in the era of big data. This paper focuses on the three environments of diversification of stakeholders, risk media, and (...) big data era. The evolution characteristics of the social risk of environmental damage of large-scale construction projects are analyzed from the four stages of incubation, outbreak, mitigation, and regression in essence. On this basis, a social risk early warning model is constructed， and the multicenter network governance mode of social risk of environmental damage in large-scale construction projects and practical social risk prevention strategies in different stages are put forward. Experiments show that the long short-term memory neural network model is effective and feasible for predicting the social risk trend of environmental damage of large-scale construction projects. Compared with other classical models, the long short-term memory model has the advantages of strong processing capability and high early warning accuracy for time-sensitive data and will have broad application prospects in the field of risk control research. By using the network governance framework and long short-term memory model, this paper studies the environmental mass events of large-scale construction projects on the risk early warning method, providing reference for the government to effectively prevent and control social risk of environmental damage of large-scale construction project in China. (shrink)

Shamans of the Piaroa ethnic group (southern Venezuela) conceive of power in terms of knowledge derived from visionary experiences. Märipa is an epistemology concerning the translation of knowledge derived from the integrative mode of consciousness, induced primarily through the consumption of plant hallucinogens, to practical effect during waking life. I integrate mythological, neurobiological, experiential, and ethnographic data to demonstrate what märipa is, and how it functions. The theory and method of märipa underlie not only Piaroa shamanic activity, but all aspects (...) of Piaroa life; mythology, causality, eschatology, and history. Piaroa shamanic practices involve conditioning the mind to achieve optimal perceptual capacities that facilitate accurate prediction and successful psycho‐social prescription. Piaroa shamans describe their technologies of consciousness in terms of gods and spirits, but also in terms of studying and the acquisition of information. Because neurobiological processes underlie the development and experience of märipa, the language of neurobiology enables a partial translation of this indigenous epistemology. The concepts of feed forward neural processing and somatic markers are central to the processes of mental imagery cultivation that Piaroa shamans employ to divine solutions to adaptive problems. Piaroa 'techniques of ecstasy' involve the ability to apply mythological templates of human adaptation to schemas of human behaviour based on years of social analysis in association with heightened information processing capacities derivative of refined experimentation with the integrative mode of consciousness. Keyordsshamanism, neurophenomenology inugrativeconsciousness, hallucinogens. (shrink)

Although great progress in neuroanatomy and physiology has occurred lately, we still cannot go directly to those levels to discover the neural mechanisms of higher cognition and consciousness. But we can use neurocomputational methods based on these details to push this project forward. Here we describe vector subtraction as an operation that computes sequential paths through high-dimensional vector spaces. Vector-space interpretations of network activity patterns are a fruitful resource in recent computational neuroscience. Vector subtraction also appears to (...) be implemented neurally in primate frontal eye field activity, which computes dimensions of saccadic eye movements. We use this apparent neural implementation as a model and construct from it a general neurocomputational account of an important type of sequential cognitive and conscious process. We defend the biological plausibility of all components of the general model and show that it yields testable anatomical and physiological predictions. We close by suggesting some interesting consequences for consciousness if our model characterizes correctly the neural mechanisms producing a common type of episode in our conscious streams. (shrink)

The detection and categorization of animate motions is a crucial task underlying social interaction and perceptual decision making. Neural representations of perceived animate objects are partially located in the primate cortical region STS, which is a region that receives convergent input from intermediate-level form and motion representations. Populations of STS cells exist which are selectively responsive to specific animated motion sequences, such as walkers. It is still unclear how and to what extent form and motion information contribute to the (...) generation of such representations and what kind of mechanisms are involved in the learning processes. The article develops a cortical model architecture for the unsupervised learning of animated motion sequence representations. We demonstrate how the model automatically selects significant motion patterns as well as meaningful static form prototypes characterized by a high degree of articulation. Such key poses are selectively reinforced during learning through a cross talk between the motion and form processing streams. Furthermore, we show how sequence-selective representations are learned in STS by fusing static form and motion input from the segregated bottom-up driving input streams. Cells in STS, in turn, feed their activities recurrently to their input sites along top-down signal pathways. We show how such learned feedback connections enable predictions about future input as anticipation generated by sequence-selective STS cells. Network simulations demonstrate the computational capacity of the proposed model by reproducing several experimental findings from neurosciences and by accounting for recent behavioral data. (shrink)

Three decades ago, William Ramsey, Steven Stich & Joseph Garon put forward an argument in favor of the following conditional: if connectionist models that implement parallelly distributed processing represent faithfully human cognitive processing, eliminativism about propositional attitudes is true. The corollary of their argument (if it proves to be sound) is that there is no place for folk psychology in contemporary cognitive science. This understanding of connectionism as a hypothesis about cognitive architecture compatible with eliminativism is also endorsed by (...) Paul Churchland, a radical opponent of folk psychology and a prominent supporter of eliminative materialism. I aim to examine whether current connectionist models based on long-short term memory (LSTM) neural networks can back up these arguments in favor of eliminativism. Nonetheless, I will rather put my faith in the eliminativism of the limited domain. This position amounts to the claim that even though connectionist cognitive science has no need whatsoever for folk psychology qua theory, this does not entail the illegitimacy of folk psychology per se in other scientific domains, most notably in humanities, but only on condition that one regards folk psychology as mere heuristics. (shrink)

Interpretation is the process whereby a hearer reasons to an interpretation of a speaker's discourse. The hearer normally adopts a credulous attitude to the discourse, at least for the purposes of interpreting it. That is to say the hearer tries to accommodate the truth of all the speaker's utterances in deriving an intended model. We present a nonmonotonic logical model of this process which defines unique minimal preferred models and efficiently simulates a kind of closed-world reasoning of particular interest for (...) human cognition. Byrne's “suppression” data (Byrne, 1989) are used to illustrate how variants on this logic can capture and motivate subtly different interpretative stances which different subjects adopt, thus indicating where more fine-grained empirical data are required to understand what subjects are doing in this task. We then show that this logical competence model can be implemented in spreading activation network models. A one pass process interprets the textual input by constructing a network which then computes minimal preferred models for (3-valued) valuations of the set of propositions of the text. The neural implementation distinguishes easy forward reasoning from more complex backward reasoning in a way that may be useful in explaining directionality in human reasoning. (shrink)

The recent discovery of the medullary circuit driving “hunger responses” – reduced thermogenesis and promoted feeding – has greatly expanded our knowledge on the central neural networks for energy homeostasis. However, how hypothalamic hunger and satiety signals generated under fasted and fed conditions, respectively, control the medullary autonomic and somatic motor mechanisms remains unknown. Here, in reviewing this field, we propose two hypothalamomedullary neural pathways for hunger and satiety signaling. To trigger hunger signaling, neuropeptide Y activates a group (...) of neurons in the paraventricular hypothalamic nucleus (PVH), which then stimulate an excitatory pathway to the medullary circuit to drive the hunger responses. In contrast, melanocortin‐mediated satiety signaling activates a distinct group of PVH neurons, which then stimulate a putatively inhibitory pathway to the medullary circuit to counteract the hunger signaling. The medullary circuit likely contains inhibitory and excitatory premotor neurons whose alternate phasic activation generates the coordinated masticatory motor rhythms to promote feeding. (shrink)

There has been much interest in the possibility of connectionist models whose representations can be endowed with compositional structure, and a variety of such models have been proposed. These models typically use distributed representations that arise from the functional composition of constituent parts. Functional composition and decomposition alone, however, yield only an implementation of classical symbolic theories. This paper explores the possibility of moving beyond implementation by exploiting holistic structure-sensitive operations on distributed representations. An experiment is performed using Pollack’s Recursive (...) Auto-Associative Memory. RAAM is used to construct distributed representations of syntactically structured sentences. A feed-forward network is then trained to operate directly on these representations, modeling syn- tactic transformations of the represented sentences. Successful training and generalization is obtained, demonstrating that the implicit structure present in these representations can be used for a kind of structure-sensitive processing unique to the connectionist domain. (shrink)

Uses connectionism (neural networks) to extract the "gist" of a story in order to represent a context going forward for the disambiguation of incoming words as a text is processed.

Human decision making shows systematic simplifications and deviations from the tenets of rationality (‘heuristics’) that may lead to suboptimal decisional outcomes (‘cognitive biases’). There are currently three prevailing theoretical perspectives on the origin of heuristics and cognitive biases: a cognitive-psychological, an ecological and an evolutionary perspective. However, these perspectives are mainly descriptive and none of them provides an overall explanatory framework for the underlying mechanisms of cognitive biases. To enhance our understanding of cognitive heuristics and biases we propose a (...) class='Hi'>neural network framework for cognitive biases, which explains why our brain systematically tends to default to heuristic (‘Type 1’) decision making. We argue that many cognitive biases arise from intrinsic brain mechanisms that are fundamental for the working of biological neural networks. In order to substantiate our viewpoint, we discern and explain four basic neural network principles: (1) Association, (2) Compatibility (3) Retainment, and (4) Focus. These principles are inherent to (all) neural networks which were originally optimized to perform concrete biological, perceptual, and motor functions. They form the basis for our inclinations to associate and combine (unrelated) information, to prioritize information that is compatible with our present state (such as knowledge, opinions and expectations), to retain given information that sometimes could better be ignored, and to focus on dominant information while ignoring relevant information that is not directly activated. The supposed mechanisms are complementary and not mutually exclusive. For different cognitive biases they may all contribute in varying degrees to distortion of information. The present viewpoint not only complements the earlier three viewpoints, but also provides a unifying and binding framework for many cognitive bias phenomena. (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.

In this paper an Artificial Neural Network (ANN) model was used to help cars dealers recognize the many characteristics of cars, including manufacturers, their location and classification of cars according to several categories including: Make, Model, Type, Origin, DriveTrain, MSRP, Invoice, EngineSize, Cylinders, Horsepower, MPG_Highway, Weight, Wheelbase, Length. ANN was used in prediction of the number of miles per gallon when the car is driven in the city(MPG_City). The results showed that ANN model was able to predict MPG_City (...) with 97.50 % accuracy. The factor of DriveTrain has the most influence on MPG_City evaluation. Similar studies can be carried out for the evaluation of other characteristics of cars. (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)

Research on natural and artificial brains is proceeding at a rapid pace. However, the understanding of the essence of consciousness has changed slightly over the millennia, and only the last decade has brought some progress to the area. Scientific ideas emerged that the soul could be a product of the material body and that calculating machines could imitate brain processes. However, the authors of this book reject the previously common dualism—the view that the material and spiritual-psychic processes are separate and (...) require a completely different substance as their foundation. -/- Reductive Model of the Conscious Mind is a forward-thinking book wherein the authors identify processes that are the essence of conscious thinking and place them in the imagined, simplified structure of cells able to memorize and transmit information in the form of impulses, which they call neurons. The purpose of the study is to explain the essence of consciousness to the degree of development of natural sciences, because only the latter can find a way to embed the concept of the conscious mind in material brains. The book is divided into three parts. Part 1 works to convince readers that the emergence of consciousness does not require detailed knowledge of the structure and morphology of the brain, with the exception of some specific properties of the neural network structure that the authors attempt to point out. Part 2 proves that the biological structure of many natural brains fulfills the necessary conditions for consciousness and intelligent thinking. Similarly, Part 3 shows the ways in which artificial creatures imitating natural brains can meet these conditions, which gives great hopes for building artificially intelligent beings endowed with consciousness. Covering topics that include cognitive architecture, the embodied mind, and machine learning, this book is ideal for cognitive scientists, philosophers of mind, neuroscientists, psychologists, researchers, academicians, and advanced-level students. The book can also help to focus the research of linguists, neurologists, and biophysicists on the biophysical basis of postulated information processing into knowledge structures. (shrink)

A serious crisis is identified in theories of neurocomputation, marked by a persistent disparity between the phenomenological or experiential account of visual perception and the neurophysiological level of description of the visual system. In particular, conventional concepts of neural processing offer no explanation for the holistic global aspects of perception identified by Gestalt theory. The problem is paradigmatic and can be traced to contemporary concepts of the functional role of the neural cell, known as the Neuron Doctrine. In (...) the absence of an alternative neurophysiologically plausible model, I propose a perceptual modeling approach, to model the percept as experienced subjectively, rather than modeling the objective neurophysiological state of the visual system that supposedly subserves that experience. A Gestalt Bubble model is presented to demonstrate how the elusive Gestalt principles of emergence, reification, and invariance can be expressed in a quantitative model of the subjective experience of visual consciousness. That model in turn reveals a unique computational strategy underlying visual processing, which is unlike any algorithm devised by man, and certainly unlike the atomistic feed-forward model of neurocomputation offered by the Neuron Doctrine paradigm. The perceptual modeling approach reveals the primary function of perception as that of generating a fully spatial virtual-reality replica of the external world in an internal representation. The common objections to this picture-in-the-head concept of perceptual representation are shown to be ill founded. Key Words: brain-anchored; Cartesian theatre; consciousness; emergence; extrinsic constraints; filling-in; Gestalt; homunculus; indirect realism; intrinsic constraints; invariance; isomorphism; multistability; objective phenomenology; perceptual modeling; perspective; phenomenology; psychophysical parallelism; psychophysical postulate; qualia; reification; representationalism; structural coherence. (shrink)

The notion of “hierarchy” is one of the most commonly posited organizational principles in systems neuroscience. To this date, however, it has received little philosophical analysis. This is unfortunate, because the general concept of hierarchy ranges over two approaches with distinct empirical commitments, and whose conceptual relations remain unclear. We call the first approach the “representational hierarchy” view, which posits that an anatomical hierarchy of feed-forward, feed-back, and lateral connections underlies a signal processing hierarchy of input-output relations. (...) Because the representational hierarchy view holds that unimodal sensory representations are subsequently elaborated into more categorical and rule-based ones, it is committed to an increasing degree of abstraction along the hierarchy. The second view, which we call “topological hierarchy", is not committed to different representational functions or degrees of abstraction at different levels. Topological approaches instead posit that the hierarchical level of a part of the brain depends on how central it is to the pattern of connections in the system. Based on the current evidence, we argue that three conceptual relations between the two approaches are possible: topological hierarchies could substantiate the traditional representational hierarchy, conflict with it, or contribute to a plurality of approaches needed to understand the organization of the brain. By articulating each of these possibilities, our analysis attempts to open a conceptual space in which further neuroscientific and philosophical reasoning about neural hierarchy can proceed. (shrink)

The topics treated in The brain and emotion include the definition, nature, and functions of emotion (Ch. 3); the neural bases of emotion (Ch. 4); reward, punishment, and emotion in brain design (Ch. 10); a theory of consciousness and its application to understanding emotion and pleasure (Ch. 9); and neural networks and emotion-related learning (Appendix). The approach is that emotions can be considered as states elicited by reinforcers (rewards and punishers). This approach helps with understanding the functions of (...) emotion, with classifying different emotions, and in understanding what information-processing systems in the brain are involved in emotion, and how they are involved. The hypothesis is developed that brains are designed around reward-and punishment-evaluation systems, because this is the way that genes can build a complex system that will produce appropriate but flexible behavior to increase fitness (Ch. 10). By specifying goals rather than particular behavioral patterns of responses, genes leave much more open the possible behavioral strategies that might be required to increase fitness. The importance of reward and punishment systems in brain design also provides a basis for understanding the brain mechanisms of motivation, as described in Chapters 2 for appetite and feeding, 5 for brain-stimulation reward, 6 for addiction, 7 for thirst, and 8 for sexual behavior. Key Words: amygdala; brain evolution; consciousness; dopamine; emotion; hunger; orbitofrontal cortex; punishment; reward; taste. (shrink)

On several key issues we agree with the commentators. Perhaps most importantly, everyone seems to agree that psychology has an important role to play in building better models of human vision, and (most) everyone agrees (including us) that deep neural networks (DNNs) will play an important role in modelling human vision going forward. But there are also disagreements about what models are for, how DNN–human correspondences should be evaluated, the value of alternative modelling approaches, and impact of marketing (...) hype in the literature. In our view, these latter issues are contributing to many unjustified claims regarding DNN–human correspondences in vision and other domains of cognition. We explore all these issues in this response. (shrink)

Bowers et al. bring forward critical issues in the current use of deep neural networks (DNNs) to model primate vision. Our own research further reveals fundamentally different algorithms utilized by DNNs for visual processing compared to the brain. It is time to reemphasize the value of basic vision research and put more resources and effort on understanding the primate brain itself.

I address whether neural networks perform computations in the sense of computability theory and computer science. I explicate and defend
the following theses. (1) Many neural networks compute—they perform computations. (2) Some neural networks compute in a classical way.
Ordinary digital computers, which are very large networks of logic gates, belong in this class of neural networks. (3) Other neural networks
compute in a non-classical way. (4) Yet other neural networks do not perform computations. Brains may well (...) fall into this last class. (shrink)

The paper aims to discuss the results of testing a neural network which classifies the vowels of the vocalic system with the [ATR] (Advanced Tongue Root) contrast based on the data of Akebu (Kwa family). The acoustic nature of the [ATR] feature is yet understudied. The only reliable acoustic correlate of [ATR] is the magnitude of the first formant (F1) which can be also modulated by tongue height, resulting in significant overlap between high [-ATR] vowels and mid [+ATR] (...) vowels. Other acoustic metrics which had been associated with the [ATR], such as F1 bandwidth (B1), relative intensity of F1 to F2 (A1-A2), etc., are typically inconsistent across vowel types and speakers. The values of four metrics – F1, F2, A1-A2, B1 – were used for training and testing the neural network. We tested four versions of the model differing in the presence of the fifth variable encoding the speaker and the number of hidden layers. The models which included the variable encoding the speaker achieved slightly higher accuracy, whereas the precision and recall metrics of the three-layer model were generally higher than those with two hidden layers. (shrink)

The paper aims to discuss the results of testing a neural network which classifies the vowels of the vocalic system with the [ATR] (Advanced Tongue Root) contrast based on the data of Akebu (Kwa family). The acoustic nature of the [ATR] feature is yet understudied. The only reliable acoustic correlate of [ATR] is the magnitude of the first formant (F1) which can be also modulated by tongue height, resulting in significant overlap between high [-ATR] vowels and mid [+ATR] (...) vowels. Other acoustic metrics which had been associated with the [ATR], such as F1 bandwidth (B1), relative intensity of F1 to F2 (A1-A2), etc., are typically inconsistent across vowel types and speakers. The values of four metrics – F1, F2, A1-A2, B1 – were used for training and testing the neural network. We tested four versions of the model differing in the presence of the fifth variable encoding the speaker and the number of hidden layers. The models which included the variable encoding the speaker achieved slightly higher accuracy, whereas the precision and recall metrics of the three-layer model were generally higher than those with two hidden layers. (shrink)

The topic of this dissertation is the nature of linguistic competence, the capacity to understand and produce sentences of natural language. I defend the empiricist account of linguistic competence embedded in the connectionist cognitive science. This strand of cognitive science has been opposed to the traditional symbolic cognitive science, coupled with transformational-generative grammar, which was committed to nativism due to the view that human cognition, including language capacity, should be construed in terms of symbolic representations and hardwired rules. Similarly, linguistic (...) competence in this framework was regarded as being innate, rule-governed, domain-specific, and fundamentally different from performance, i.e., idiosyncrasies and factors governing linguistic behavior. I analyze state-of-the-art connectionist, deep learning models of natural language processing, most notably large language models, to see what they can tell us about linguistic competence. Deep learning is a statistical technique for the classification of patterns through which artificial intelligence researchers train artificial neural networks containing multiple layers that crunch a gargantuan amount of textual and/or visual data. I argue that these models suggest that linguistic competence should be construed as stochastic, pattern-based, and stemming from domain-general mechanisms. Moreover, I distinguish syntactic from semantic competence, and I show for each the ramifications of the endorsement of a connectionist research program as opposed to the traditional symbolic cognitive science and transformational-generative grammar. I provide a unifying front, consisting of usage-based theories, a construction grammar approach, and an embodied approach to cognition to show that the more multimodal and diverse models are in terms of architectural features and training data, the stronger the case is for the connectionist linguistic competence. I also propose to discard the competence vs. performance distinction as theoretically inferior so that a novel and integrative account of linguistic competence originating in connectionism and empiricism that I propose and defend in the dissertation could be put forward in scientific and philosophical literature. (shrink)

The Air Quality Index (AQI) shows the state of air pollution in a unique and more understandable way. This work aims to forecast the AQI in Algeciras (Spain) 8 hours in advance. The AQI is calculated indirectly through the predicted concentrations of five pollutants (O3, NO2, CO, SO2 and PM10) to achieve this goal. Artificial neural networks (ANNs), sequence-to-sequence long short-term memory networks (LSTMs) and a newly proposed method combing a rolling window with the latter (LSTMNA) are employed as (...) the forecasting techniques. Besides, two input approaches are evaluated: using only the data from the own time series of the pollutant in the first case or adding exogenous variables in the second one. Several window sizes are employed (24, 28 and 72 hours) with ANNs and LSTMNAs. Additionally, several feature ranking methods are applied in the exogenous approach to select the most relevant lagged variables to feed the models. Results show how the proposed exogenous approach increases the performance of the prediction models. Besides, the newly proposed method LSTMNA provides the best performances in most of the cases evaluated. Hence, it constitutes an exciting alternative to standard LSTMs and ANNs to predict pollutants concentrations and, consequently, the AQI. (shrink)

Bowers et al. counter deep neural networks (DNNs) as good models of human visual perception. From our color perspective we feel their view is based on three misconceptions: A misrepresentation of the state-of-the-art of color perception; the type of model required to move the field forward; and the attribution of shortcomings to DNN research that are already being resolved.

Interpretation is the process whereby a hearer reasons to an interpretation of a speaker's discourse. The hearer normally adopts a credulous attitude to the discourse, at least for the purposes of interpreting it. That is to say the hearer tries to accommodate the truth of all the speaker's utterances in deriving an intended model. We present a nonmonotonic logical model of this process which defines unique minimal preferred models and efficiently simulates a kind of closed-world reasoning of particular interest for (...) human cognition. Byrne's “suppression” data (Byrne, 1989) are used to illustrate how variants on this logic can capture and motivate subtly different interpretative stances which different subjects adopt, thus indicating where more fine-grained empirical data are required to understand what subjects are doing in this task. We then show that this logical competence model can be implemented in spreading activation network models. A one pass process interprets the textual input by constructing a network which then computes minimal preferred models for (3-valued) valuations of the set of propositions of the text. The neural implementation distinguishes easy forward reasoning from more complex backward reasoning in a way that may be useful in explaining directionality in human reasoning. (shrink)