Hebbian learning

Musician's brains constitute an interesting model for neuroplasticity. Imaging studies demonstrated that sensorimotor cortical representations are altered in musicians, which was assumed to arise from the development of skilled performance. However, the perceptual consequences of ...

The goal of this article is to propose a new cognitive model that focuses on bottom-up learning of explicit knowledge (i.e., the transformation of implicit knowledge into explicit knowledge). This phenomenon has recently received much attention in empirical research that was not accompanied by a corresponding work effort in cognitive modeling. The new model is called TEnsor LEarning of CAusal STructure (TELECAST). In TELECAST, implicit processing is modeled using an unsupervised connectionist network (the Joint Probability EXtractor: JPEX) while explicit (causal) knowledge is implemented using a Bayesian belief network (which is built online using JPEX). Every task is simultaneously processed explicitly and implicitly and the results are integrated to provide the model output. Here, TELECAST is used to simulate a causal inference task and two serial reaction time experiments.

This thesis provides a theoretical description of on-line unsupervised learning from high-dimensional data. In particular, the learning dynamics of the on-line Hebbian algorithm is studied for the following two popular statistical models—principal component analysis (PCA) and independent component analysis (ICA). The methods of statistical mechanics are used to elucidate the critical transient stages of the learning process.

In the case of on-line PCA, the statistical mechanics approach allows the derivation of a set of deterministic differential equations for a small number of macroscopic self-averaging order parameters which enables an exact calculation of the evolution of the error function in the limit of large data dimension. In the case of ICA learning, it is found that the interesting macroscopic order parameters are not self-averaging in general and the transient dynamics is relatively slow with large fluctuations. The parameter trajectory of the on-line Hebbian ICA algorithm studied here passes through a sequence of metastable states, each of which can be described by a diffusion process in a polynomial potential. The proper treatment of the fluctuations of the order parameters for this case is given by the Fokker-Planck equation.

For both models a natural gradient version of the studied algorithms is constructed using the fact that the parameter space of our models is constrained to a Stiefel manifold of orthogonal rectangular matrices. In the case of on-line ICA, the natural gradient variant exhibits the same stochastic trapping in sub-optimal metastable states during transient learning as in the case of the standard Hebbian ICA algorithm and there is only an advantage in using the natural gradient variant asymptotically.

The problem of the sensitivity of on-line learning to the choice of learning algorithms is considered. Recommendations for proper adjustment of learning parameters in order to improve the performance of studied algorithms are given. Numerical simulations for finite-sized systems are in good agreement with the theoretical results.

The process of segmenting images is one of the most critical ones in automatic image analysis whose goal can be regarded as to find what objects are present in images. Artificial neural networks have been well developed so far. First two generations of neural networks have a lot of successful applications. Spiking neuron networks (SNNs) are often referred to as the third generation of neural networks which have potential to solve problems related to biological stimuli. They derive their strength and interest from an accurate modeling of synaptic interactions between neurons, taking into account the time of spike emission. SNNs overcome the computational power of neural networks made of threshold or sigmoidal units. Based on dynamic event-driven processing, they open up new horizons for developing models with an exponential capacity of memorizing and a strong ability to fast adaptation. Moreover, SNNs add a new dimension, the temporal axis, to the representation capacity and the processing abilities of neural networks. In this paper, we present how SNN can be applied with efficacy in image segmentation and edge detection. Results obtained confirm the validity of the approach.

Hebbian learning is a biologically plausible and ecologically valid learning mechanism. In Hebbian learning, 'units that fire together, wire together'. Such learning may occur at the neural level in terms of long-term potentiation (LTP) and long-term depression (LTD). Many features of Hebbian learning are relevant to developmental theorizing, including its self-organizing nature and its ability to extract statistical regularities from the environment. Hebbian learning mechanisms may also play an important role in critical periods during development, and in a number of other developmental phenomena.

Computational models in cognitive neuroscience should ideally use biological properties and powerful computational principles to produce behavior consistent with psychological findings. Error-driven backpropagation is computationally powerful, and has proven useful for modeling a range of psychological data, but is not biologically plausible. Several approaches to implementing backpropagation in a biologically plausible fashion converge on the idea of using bidirectional activation propagation in interactive networks to convey error signals. This paper demonstrates two main points about these error-driven interactive networks: (a) they generalize poorly due to attractor dynamics that interfere with the network's ability to systematically produce novel combinatorial representations in response to novel inputs; and (b) this generalization problem can be remedied by adding two widely used mechanistic principles, inhibitory competition and Hebbian learning, that can be independently motivated for a variety of biological, psychological and computational reasons. Simulations using the Leabra algorithm, which combines the generalized recirculation (GeneRec) biologically-plausible error-driven learning algorithm with inhibitory competition and Hebbian learning, show that these mechanisms can result in good generalization in interactive networks. These results support the general conclusion that cognitive neuroscience models that incorporate the core mechanistic principles of interactivity, inhibitory competition, and error-driven and Hebbian learning satisfy a wider range of biological, psychological and computational constraints than models employing a subset of these principles.

The error backpropagation learning algorithm (BP) is generally considered biologically implausible because it does not use locally available, activation-based variables. A version of BP that can be computed locally using bi-directional activation recirculation instead of backpropagated error derivatives is more biologically plausible. This paper presents a generalized version of the recirculation algorithm (GeneRec), which overcomes several limitations of the earlier algorithm by using a generic recurrent network with sigmoidal units that can learn arbitrary input/output mappings. However, the contrastive-Hebbian learning algorithm (CHL, a.k.a. DBM or mean field learning) also uses local variables to perform error-driven learning in a sigmoidal recurrent network. CH L was derived in a stochastic framework (the Boltzmann machine), but has been extended to the deterministic case in various ways, all of which rely on problematic approximations and assumptions, leading some to conclude that it is fundamentally flawed. This paper shows that CHL can be derived instead from within the BP framework via the GeneRec algorithm. CHL is a symmetry-preserving version of GeneRec which uses a simple approximation to the midpoint or second-order accurate Runge-Kutta method of numerical integration, which explains the generally faster learning speed of CHL compared to BP. Thus, all known fully general error-driven learning algorithms that use local activation-based variables in deterministic networks can be considered variations of the GeneRec algorithm (and indirectly, of the backpropagation algorithm). GeneRec therefore provides a promising framework for thinking about how the brain might perform error-driven learning. To further this goal, an explicit biological mechanism is proposed which would be capable of implementing GeneRec-style learning. This mechanism is consistent with available evidence regarding synaptic modification in neurons in the neocortex and hippocampus, and makes further predictions.

Abstract—Spiking neural network (NN) architecture that uses Hebbian learning and reinforcement-learning schemes for adapting the synaptic weights is implemented in silicon and performs dynamic optimization according to hemodynamic sensor for a cardiac resynchronization therapy (CRT) device. The spiking NN architecture dynamically changes the atrioventricular (AV) delay and interventricular (VV) interval parameters according to the information provided by the intracardiac electrograms (IEGMs) and hemodynamic sensors. The spiking NN coprocessor performs the adaptive part and is controlled by a deterministic algorithm master controller. The simulated cardiac output obtained with the adaptive CRT device is 30 % higher than with a nonadaptive CRT device and is likely to provide improvement in the quality of life for patients with congestive heart failure. The spiking NN architecture shows synaptic plasticity acquired during the learning process. The synaptic plasticity is manifested by a...

There is a convergence between cognitive models of imitation, constructs derived from social psychology studies on mimicry and empathy, and recent empirical findings from the neurosciences. The ideomotor framework of human actions assumes a common representational format for action and perception that facilitates imitation. Furthermore, the associative sequence learning model of imitation proposes that experience-based Hebbian learning forms links between sensory processing of the actions of others and motor plans. Social psychology studies have demonstrated that imitation and mimicry are pervasive, automatic, and facilitate empathy. Neuroscience investigations have demonstrated physiological mechanisms of mirroring at single-cell and neural-system levels that support the cognitive and social psychology constructs. Why were these neural mechanisms selected, and what is their adaptive advantage? Neural mirroring solves the "problem of other minds" (how we can access and understand the minds of others) and makes intersubjectivity possible, thus facilitating social behavior. 653 Annu. Rev. Psychol. 2009.60:653-670. Downloaded from arjournals.annualreviews.org by Hebrew University of Jerusalem on 12/29/08. For personal use only.

How does the brain form a useful representation of its environment? It is shown here that a layer of simple Hebbian units connected by modifiable anti-Hebbian feed-back connections can learn to code a set of patterns in such a way that statistical dependency between the elements of the representation is reduced, while information is preserved. The resulting code is sparse, which is favourable if it is to be used as input to a subsequent supervised associative layer. The operation of the network is demonstrated on two simple problems.

a b d This paper focuses on the issue of developing self-adapting automatic object detection systems for improving their performance. ' b o general methodologies for performance improvement are first introduced. They are based on parameter optimizing and input adapting. Different modified Hebbian learning rules are developed to build adaptive feature extractors which transform the input data into a desired form for a given algorithm. To show its feasibility, an input adaptor for object detection is designed as an example and tested using multisensor data (optical, SAR, and FLIR). Test results are presented and discussed in the paper.

The learning dynamics of on-line independent component analysis is analysed in the limit of large data dimension. We study a simple Hebbian learning algorithm that can be used to separate out a small number of non-Gaussian components from a high-dimensional data set. The de-mixing matrix parameters are confined to a Stiefel manifold of tall, orthogonal matrices and we introduce a natural gradient variant of the algorithm which is appropriate to learning on this manifold. For large input dimension the parameter trajectory of both algorithms passes through a sequence of unstable fixed points, each described by a diffusion process in a polynomial potential. Choosing the learning rate too large increases the escape time from each of these fixed points, effectively trapping the learning in a sub-optimal state. In order to avoid these trapping states a very low learning rate must be chosen during the learning transient, resulting in learning time-scales of O(N^2) or O(N^3) iterations where N is the data dimension. Escape from each sub-optimal state results in a sequence of symmetry breaking events as the algorithm learns each source in turn. This is in marked contrast to the learning dynamics displayed by related on-line learning algorithms for multilayer neural networks and principal component analysis. Although the natural gradient variant of the algorithm has nice asymptotic convergence properties, it has an equivalent transient dynamics to the standard Hebbian algorithm.

Autoencoders are unsupervised machine learning circuits, with typically one hidden layer, whose learning goal is to minimize an average distortion measure between inputs and outputs. Linear autoencoders correspond to the special case where only linear transformations between visible and hidden variables are used. While linear autoencoders can be defined over any field, only real-valued linear autoencoders have been studied so far. Here we study complex-valued linear autoencoders where the components of the training vectors and adjustable matrices are defined over the complex field with the L 2 norm. We provide simpler and more general proofs that unify the realvalued and complex-valued cases, showing that in both cases the landscape of the error function is invariant under certain groups of transformations. The landscape has no local minima, a family of global minima associated with Principal Component Analysis, and many families of saddle points associated with orthogonal projections onto sub-space spanned by sub-optimal subsets of eigenvectors of the covariance matrix. The theory yields several iterative, convergent, learning algorithms, a clear understanding of the generalization properties of the trained autoencoders, and can equally be applied to the hetero-associative case when external targets are provided. Partial results on deep architecture as well as the differential geometry of autoencoders are also presented. The general framework described here is useful to classify autoencoders and identify general properties that ought to be investigated for each class, illuminating some of the connections between autoencoders, unsupervised learning, clustering, Hebbian learning, and information theory.

This study presents a survey on the most recent learning approaches and algorithms that are related to fuzzy cognitive maps (FCMs). FCMs are cognition fuzzy influence graphs, which are based on fuzzy logic and neural network aspects that inherit their main advantages. They gained momentum due to their dynamic characteristics and learning capabilities. These capabilities make them essential for modeling and decision-making tasks as they improve the performance of these tasks. An efficient number of learning algorithms for FCMs, by modifying the FCM weight matrix, have been developed in order to update the initial knowledge of human experts and/or include any knowledge from historical data in order to produce learned weights. The proposed learning techniques have mainly been concentrated on three directions: on the production of weight matrices on the basis of historical data, on adaptation of the cause-effect relationships of the FCM on the basis of experts' intervention, and on the production of weight matrices by combining experts' knowledge and data. The learning techniques could be categorized into three groups on the basis of the learning paradigm: Hebbian-based, populationbased, and hybrid, which subsequently combine the main aspects of Hebbian-based-and population-based-type learning algorithms. These types of learning algorithms are the most efficient and widely used to train the FCMs, according to the existing literature. A survey on recent advances on learning methodologies and algorithms for FCMs that present their dynamic capabilities and application characteristics in diverse scientific fields is established here.

In this work we investigate the possibilities offered by a minimal framework of artificial spiking neurons to be deployed in silico. Here we introduce a hierarchical network architecture of spiking neurons which learns to recognize moving objects in a visual environment and determine the correct motor output for each object. These tasks are learned through both supervised and unsupervised spike timing dependent plasticity (STDP). STDP is responsible for the strengthening (or weakening) of synapses in relation to pre- and post-synaptic spike times and has been described as a Hebbian paradigm taking place both in vitro and in vivo. We utilize a variation of STDP learning, called burst-STDP, which is based on the notion that, since spikes are expensive in terms of energy consumption, then strong bursting activity carries more information than single (sparse) spikes. Furthermore, this learning algorithm takes advantage of homeostatic renormalization, which has been hypothesized to promote memory consolidation during NREM sleep. Using this learning rule, we design a spiking neural network architecture capable of object recognition, motion detection, attention towards important objects, and motor control outputs. We demonstrate the abilities of our design in a simple environment with distractor objects, multiple objects moving concurrently, and in the presence of noise. Most importantly, we show how this neural network is capable of performing these tasks using a simple leaky-integrate-and-fire (LIF) neuron model with binary synapses, making it fully compatible with state-of-the-art digital neuromorphic hardware designs. As such, the building blocks and learning rules presented in this paper appear promising for scalable fully neuromorphic systems to be implemented in hardware chips.

Fuzzy cognitive maps have gained considerable research interest and widely used to analyze complex systems and making decisions. Recently they have been found large applicability in diverse domains for decision support and classification tasks. A new learning paradigm for FCMs is proposed in this research work, inheriting the main aspects of ensemble based learning approaches, such as bagging and boosting. FCM ensemble learning is an approach where the model is trained using non linear Hebbian learning (NHL) algorithm and further its performance is enhanced using ensemble techniques. This work is inspired from the neural networks ensembles and it is used to learn the FCMs ensembles produced by the already known and efficient data driven NHL algorithm. The new proposed approach of FCM ensembles is applied to identification of Autism and the results are compared with those produced by data driven NHL algorithm alone for FCM training. Experimental results demonstrate that the proposed FCM ensemble algorithm works better than the NHL-based approach alone with respect to accuracy for learning FCM.

We study inverse problems in neural field theory, i.e., the construction of synaptic weight kernels yielding a prescribed neural field dynamics. We address the issues of existence, uniqueness, and stability of solutions to the inverse problem for the Amari neural field equation as a special case, and prove that these problems are generally ill-posed. In order to construct solutions to the inverse problem, we first recast the Amari equation into a linear perceptron equation in an infinitedimensional Banach or Hilbert space. In a second step, we construct sets of biorthogonal function systems allowing the approximation of synaptic weight kernels by a generalized Hebbian learning rule. Numerically, this construction is implemented by the Moore-Penrose pseudoinverse method. We demonstrate the instability of these solutions and use the Tikhonov regularization method for stabilization and to prevent numerical overfitting. We illustrate the stable construction of kernels by means of three instructive examples.

We present a mathematical analysis of the effects of Hebbian learning in random recurrent neural networks, with a generic Hebbian learning rule including passive forgetting and different time scales for neuronal activity and learning dynamics.

Inverse problems for dynamical system models of cognitive processes comprise the determination of synaptic weight matrices or kernel functions for neural networks or neural/dynamic field models, respectively. We introduce dynamic cognitive modeling as a three tier top-down approach where cognitive processes are first described as algorithms that operate on complex symbolic data structures. Second, symbolic expressions and operations are represented by states and transformations in abstract vector spaces. Third, prescribed trajectories through representation space are implemented in neurodynamical systems. We discuss the Amari equation for a neural/dynamic field theory as a special case and show that the kernel construction problem is particularly ill-posed. We suggest a Tikhonov-Hebbian learning method as regularization technique and demonstrate its validity and robustness for basic examples of cognitive computations.

Fuzzy Cognitive Maps (FCMs) constitute an attractive knowledge-based methodology, combining the robust properties of fuzzy logic and neural networks. FCMs represent causal knowledge as a signed directed graph with feedback and provide an intuitive framework which incorporates the experts' knowledge. FCMs handle available information and knowledge from an abstract point of view. They develop behavioural model of the system exploiting the experience and knowledge of experts. The construction of FCMs is based mainly on experts who determine the structure of FCM, i.e. concepts and weighted interconnections among concepts. But this methodology may not be a sufficient model of the system because the human factor is not always reliable. Thus the FCM model of the system may requires restructuring which is achieved through adjustment the weights of FCM interconnections using specific learning algorithms for FCMs. In this article, two unsupervised learning algorithms are presented and compared for training FCMs; how they define, select or fine-tuning weights of the causal interconnections among concepts. The implementation and results of these unsupervised learning techniques for an industrial process control problem are discussed. The simulations results of training the process system verify the effectiveness, validity and advantageous characteristics of those learning techniques for FCMs. r

Fuzzy Associative Conjuncted Maps (FASCOM) is a fuzzy neural network that represents information by conjuncting fuzzy sets and associates them through a combination of unsupervised and supervised learning. The network first quantizes input and output feature maps using fuzzy sets. They are subsequently conjuncted to form antecedents and consequences, and associated to form fuzzy if-then rules. These associations are learnt through a learning process consisting of three consecutive phases. First, an unsupervised phase initializes based on information density the fuzzy membership functions that partition each feature map. Next, a supervised Hebbian learning phase encodes synaptic weights of the input-output associations. Finally, a supervised error reduction phase fine-tunes the fine-tunes the network and discovers the varying influence of an input dimension across output feature space. FASCOM was benchmarked against other prominent architectures using data taken from three nonlinear data estimation tasks and a realworld road traffic density prediction problem. The promising results compiled show significant improvements over the stateof-the-art for all four data prediction tasks.

The firing rate of single neurons in the mammalian hippocampus has been demonstrated to encode for a range of spatial and non-spatial stimuli. It has also been demonstrated that phase of firing, with respect to the theta oscillation that dominates the hippocampal EEG during stereotype learning behaviour, correlates with an animal's spatial location. These findings have led to the hypothesis that the hippocampus operates using a dual (rate and temporal) coding system. To investigate the phenomenon of dual coding in the hippocampus, we examine a spiking recurrent network model with theta coded neural dynamics and an STDP rule that mediates rate-coded Hebbian learning when pre-and post-synaptic firing is stochastic. We demonstrate that this plasticity rule can generate both symmetric and asymmetric connections between neurons that fire at concurrent or successive theta phase, respectively, and subsequently produce both pattern completion and sequence prediction from partial cues. This unifies previously disparate auto-and hetero-associative network models of hippocampal function and provides them with a firmer basis in modern neurobiology. Furthermore, the encoding and reactivation of activity in mutually exciting Hebbian cell assemblies demonstrated here is believed to represent a fundamental mechanism of cognitive processing in the brain.

It has been suggested that information in the brain is encoded in temporal spike patterns which are decoded by a combination of time delays and coincidence detection. Here, we show how a multi-compartmental model of a cerebellar Purkinje cell can learn to recognise temporal parallel fibre activity patterns by adapting latencies of calcium responses after activation of metabotropic glutamate receptors (mGluRs). In each compartment of our model, the mGluR signalling cascade is represented by a set of differential equations that reflect the underlying biochemistry. Phosphorylation of the mGluRs changes the concentration of receptors which are available for activation by glutamate and thereby adjusts the time delay between mGluR stimulation and voltage response. The adaptation of a synaptic delay as opposed to a weight represents a novel non-Hebbian learning mechanism that can also implement the adaptive timing of the classically conditioned eye-blink response.

Meaningful familiar stimuli and senseless unknown materials lead to different patterns of brain activation. A late major neurophysiological response indexing ‘sense’ is the negative component of event-related potential peaking at around 400 ms (N400), an event-related potential that emerges in attention-demanding tasks and is larger for senseless materials (e.g. meaningless pseudowords) than for matched meaningful stimuli (words). However, the mismatch negativity (latency 100–250 ms), an early automatic brain response elicited under distraction, is larger to words than to pseudowords, thus exhibiting the opposite pattern to that seen for the N400. So far, no theoretical account has been able to reconcile and explain these findings by means of a single, mechanistic neural model. We implemented a neuroanatomically grounded neural network model of the left perisylvian language cortex and simulated: (i) brain processes of early language acquisition and (ii) cortical responses to familiar word and senseless pseudoword stimuli. We found that variation of the area-specific inhibition (the model correlate of attention) modulated the simulated brain response to words and pseudowords, producing either an N400- or a mismatch negativity-like response depending on the amount of inhibition (i.e. available attentional resources). Our model: (i) provides a unifying explanatory account, at cortical level, of experimental observations that, so far, had not been given a coherent interpretation within a single framework; (ii) demonstrates the viability of purely Hebbian, associative learning in a multilayered neural network architecture; and (iii) makes clear predictions on the effects of attention on latency and magnitude of event-related potentials to lexical items. Such predictions have been confirmed by recent experimental evidence.

are a class of densely connected single-layer nonlinear networks of perceptrons. The network's energy function is defined through a learning procedure so that its minima coincide with states from a predefined set. However, because of the network's nonlinearity, a number of undesirable local energy minima emerge from the learning procedure. This has shown to significantly effect the network's performance. In this brief, we present a stochastic process-enhanced binary HNN. Given a fixed network topology, the desired final distribution of states can be reached by modulating the network's stochastic process. We design this process, in a computationally efficient manner, by associating it with stability intervals of the nondesired stable states of the network. Our experimental simulations confirm the predicted improvement in performance.

Cognitive theory has decomposed human mental abilities into cognitive (sub) systems, and cognitive neuroscience succeeded in disclosing a host of relationships between cognitive systems and specific structures of the human brain. However, an explanation of why specific functions are located in specific brain loci had still been missing, along with a neurobiological model that makes concrete the neuronal circuits that carry thoughts and meaning. Brain theory, in particular the Hebb-inspired neurocybernetic proposals by Braitenberg, now offers an avenue toward explaining brain–mind relationships and to spell out cognition in terms of neuron circuits in a neuromechanistic sense. Central to this endeavor is the theoretical construct of an elementary functional neuronal unit above the level of individual neurons and below that of whole brain areas and systems: the distributed neuronal assembly (DNA) or thought circuit (TC). It is shown that DNA/TC theory of cognition offers an integrated explanatory perspective on brain mechanisms of perception, action, language, attention, memory, decision and conceptual thought. We argue that DNAs carry all of these functions and that their inner structure (e.g., core and halo subcomponents), and their functional activation dynamics (e.g., ignition and reverberation processes) answer crucial localist questions, such as why memory and decisions draw on prefrontal areas although memory formation is normally driven by information in the senses and in the motor system. We suggest that the ability of building DNAs/TCs spread out over different cortical areas is the key mechanism for a range of specifically human sensorimotor, linguistic and conceptual capacities and that the cell assembly mechanism of overlap reduction is crucial for differentiating a vocabulary of actions, symbols and concepts.

Recent models of the oculomotor delayed response task have been based on the assumption that working memory is stored as a persistent activity state (a 'bump' state). The delay activity is maintained by a finely tuned synaptic weight matrix producing a line attractor. Here we present an alternative hypothesis, that fast Hebbian synaptic plasticity is the mechanism underlying working memory. A computational model demonstrates a working memory function that is more resistant to distractors and network inhomogeneity compared to previous models, and that is also capable of storing multiple memories.

Learning and memory operations in neural circuits are believed to involve molecular cascades of synaptic and nonsynaptic changes that lead to a diverse repertoire of dynamical phenomena at higher levels of processing. Hebbian and homeostatic plasticity, neuromodulation, and intrinsic excitability all conspire to form and maintain memories. But it is still unclear how these seemingly redundant mechanisms could jointly orchestrate learning in a more unified system. To this end, a Hebbian learning rule for spiking neurons inspired by Bayesian statistics is proposed. In this model, synaptic weights and intrinsic currents are adapted on-line upon arrival of single spikes, which initiate a cascade of temporally interacting memory traces that locally estimate probabilities associated with relative neuronal activation levels. Trace dynamics enable synaptic learning to readily demonstrate a spike-timing dependence, stably return to a set-point over long time scales, and remain competitive despite this stability. Beyond unsupervised learning, linking the traces with an external plasticity-modulating signal enables spike-based reinforcement learning. At the postsynaptic neuron, the traces are represented by an activity-dependent ion channel that is shown to regulate the input received by a postsynaptic cell and generate intrinsic graded persistent firing levels. We show how spike-based Hebbian-Bayesian learning can be performed in a simulated inference task using integrate-and-fire (IAF) neurons that are Poisson-firing and background-driven, similar to the preferred regime of cortical neurons. Our results support the view that neurons can represent information in the form of probability distributions, and that probabilistic inference could be a functional by-product of coupled synaptic and nonsynaptic mechanisms operating over several timescales. The model provides a biophysical realization of Bayesian computation by reconciling several observed neural phenomena whose functional effects are only partially understood in concert.

Among passerines, Bengali finches are known to sing extremely complex courtship songs with three hierarchical structures: namely, the element, the chunk, and the syntax. In this work, we theoretically studied the mechanism of the song of Bengali finches in aides to provide a dynamic view of the development of birdsong learning. We first constructed a model of the Elman network with chaotic neurons that successfully learned the supervisor signal defined by a simple finite-state syntax. Second, we focused on the process of individual-specific increases in the complexity of song syntax. We propose a new learning algorithm to produce the intrinsic diversification of song syntax without a supervisor on the basis of the itinerant dynamics of chaotic neural networks and the Hebbian learning rule. The emergence of novel syntax modifying the acquired syntax is demonstrated.

Neural plasticity has been invoked as a powerful argument against nativism. However, there is a line of argument, which is well exemplified by Pinker (2002) and more recently by Laurence and Margolis (2015a) with respect to concept na-tivism, according to which even extreme cases of plasticity show important innate constraints, so that one should rather speak of " constrained plasticity ". According to this view, cortical areas are not really equipotential, they perform instead different kinds of computation, follow essentially different learning rules, or have a fixed internal structure acting as a filter for specific categories of inputs. We intend to analyze this argument, in the light of a review of current neuroscientific literature on plasticity. Our conclusion is that Laurence and Margolis are right in their appeal to innate constraints on connectivity – a thesis that is nowadays welcome to both na-tivists (Mahon and Caramazza, 2011) and non-nativists (Pulvermüller et al, 2014) – but there is little support for their claim of further innate differentiation between and within cortical areas. As we will show, there is instead strong evidence that the cortex is characterized by the indefinite repetition of substantially identical computational units, giving rise in any of its portions to Hebbian, input-dependent plasticity. Although this is entirely compatible with the existence of innate constraints on the brain's connectivity, the cerebral cortex architecture based on a multiplicity of maps correlating with one another has important computational consequences, a point that has been underestimated by traditional connectionist approaches.

In this article, we present a biophysical model of STDP based on a differential Hebbian learning rule (ISO learning). This rule correlates presynaptically the NMDA channel conductance with the derivative of the membrane potential at the synapse as the postsynaptic signal. The model is able to reproduce the generic STDP weight change characteristic. We nd that (1) The actual shape of the weight change curve strongly depends on the NMDA channel characteristics and on the shape of the membrane potential at the synapse.

In practical data mining problems high-dimensional data has to be analyzed. In most of these cases it is very informative to map and visualize the hidden structure of complex data set in a low-dimensional space. The aim of this paper is to propose a new mapping algorithm based both on the topology and the metric of the data. The utilized Topology Representing Network (TRN) combines neural gas vector quantization and competitive Hebbian learning rule in such a way that the hidden data structure is approximated by a compact graph representation. TRN is able to define a low-dimensional manifold in the high-dimensional feature space. In case the existence of a manifold, multidimensional scaling and/or Sammon mapping of the graph distances can be used to form the map of the TRN (TRNMap). The systematic analysis of the algorithms that can be used for data visualization and the numerical examples presented in this paper demonstrate that the resulting map gives a good representation of the topology and the metric of complex data sets, and the component plane representation of TRNMap is useful to explore the hidden relations among the features.

It is difficult to map many existing learning algorithms onto biological networks because the former require a separate learning network. The computational basis of biological cortical learning is still poorly understood. This paper rigorously introduces a concept called in-place learning. With in-place learning, every networked neuron in-place is responsible for the learning of its signal processing characteristics (e.g., efficacies of synapses) within its connected network environment. There is no need for a separate learning network. With this in-place hypothesis, consequently, each neuron does not have extra space to compute and store the second and higher order statistics (e.g., correlations) of its input fibers. This work first provides a classification of learning algorithms. Then, it shows that the two well-known in-place biological mechanisms, the Hebbian rule and lateral inhibition, are sufficient to develop orientation selective cells, similar to those found in V1, from inputs of natural images. Many other cells that have not been fully understood have emerged as well. The presented computational study of these two in-place learning mechanisms leads to a new concept-these cells correspond to what are called lobe components, which are high concentrations in the probability of the neuronal input space. Further analysis explains how every neuron can learn efficiently (i.e., near-optimal efficiency) by scheduling its plasticity while interacting with other connected neurons. A simple, in-place (Type-5) learning algorithm is presented. The experimental results showed that this simple biologically inspired algorithm is superior to some well-known state-of-the-art ICA algorithms, thanks to its nearoptimal efficiency.

This article based on the book « La métamorphose identitaire » (Duval, 2019) an explanatory descriptive research (Trudel, L., Simard, C., Vonarx, N., 2007) by multiple case study "... aims to draw conclusions from a set of cases" (Gagnon, 2012). The theoretical aspect stands on a documentary exploration in various fields of human and cognitive sciences. This research stems from a personal experience and questions about how an identity metamorphosis occurs. The hypothesis proposed by the present study is that: the process of identity metamorphosis would be similar to the conceptual metaphor process. We believe that the intensity of the confrontation created between the old reference frame and the reference frame containing the new knowledge would modify the structure of understanding and self-recognition creating a new mental schema, a third frame (a gestalt), that of an identity metamorphosis. An analysis of four public cases of identity metamorphosis is presented as well as its process and action sequence. Résumé: Cet article tiré du livre « La métamorphose identitaire » (Duval, 2019) est une recherche descriptive explicative (Trudel, L., Simard, C., Vonarx, N., 2007) par étude de cas multiples qui « … vise à tirer des conclusions d'un ensemble de cas » (Gagnon, 2012). L'aspect théorique a comme source une exploration documentaire dans divers domaines des sciences humaines et cognitives. Cette recherche a comme origine une expérience personnelle et une interrogation concernant le mode de fonctionnement d'une métamorphose identitaire. L'hypothèse proposée par la présente étude est que : le processus de la métamorphose identitaire serait similaire à celui de la métaphore conceptuelle. L'intensité de la confrontation créée entre l'ancien cadre de référence et le nouveau cadre contenant les nouvelles connaissances modifierait la structure de compréhension et de reconnaissance de soi créant un nouveau schéma mental, un troisième cadre (une gestalt), celui de la métamorphose identitaire. Une analyse de quatre cas publique de métamorphose identitaire est présentée de même qu'un processus et une séquence d'action.

Mirror neurons are increasingly recognized as a crucial substrate for many developmental processes, including imitation and social learning. Although there has been considerable progress in describing their function and localization in the primate and adult human brain, we still know little about their ontogeny. The idea that mirror neurons result from Hebbian learning while the child observes/hears his/her own actions has received remarkable empirical support in recent years. Here we add a new element to this proposal, by suggesting that the infant's perceptual-motor system is optimized to provide the brain with the correct input for Hebbian learning, thus facilitating the association between the perception of actions and their corresponding motor programs. We review evidence that infants (1) have a marked visual preference for hands, (2) show cyclic movement patterns with a frequency that could be in the optimal range for enhanced Hebbian learning, and (3) show synchronized theta EEG (also known to favour synaptic Hebbian learning) in mirror cortical areas during self-observation of grasping. These conditions, taken together, would allow mirror neurons for manual actions to develop quickly and reliably through experiential canalization. Our hypothesis provides a plausible pathway for the emergence of mirror neurons that integrates learning with genetic pre-programming, suggesting new avenues for research on the link between synaptic processes and behaviour in ontogeny.

A correlation-based ͑''Hebbian''͒ learning rule at a spike level with millisecond resolution is formulated, mathematically analyzed, and compared with learning in a firing-rate description. The relative timing of presynaptic and postsynaptic spikes influences synaptic weights via an asymmetric ''learning window.'' A differential equation for the learning dynamics is derived under the assumption that the time scales of learning and neuronal spike dynamics can be separated. The differential equation is solved for a Poissonian neuron model with stochastic spike arrival. It is shown that correlations between input and output spikes tend to stabilize structure formation. With an appropriate choice of parameters, learning leads to an intrinsic normalization of the average weight and the output firing rate. Noise generates diffusion-like spreading of synaptic weights.

Experimental evidence indicates that synaptic modification depends on the timing relationship between the presynaptic inputs and the output spikes that they generate. In this letter, results are presented for models of spike-timing-dependent plasticity (STDP) whose weight dynamics is determined by a stable fixed point. Four classes of STDP are identified on the basis of the time extent of their input-output interactions. The effect on the potentiation of synapses with different rates of input is investigated to elucidate the relationship of STDP with classical studies of long-term potentiation and depression and rate-based Hebbian learning. The selective potentiation of higher-rate synaptic inputs is found only for models where the time extent of the input-output interactions is input restricted (i.e., restricted to time domains delimited by adjacent synaptic inputs) and that have a time-asymmetric learning window with a longer time constant for depression than for potentiation. The analysis provides an account of learning dynamics determined by an input-selective stable fixed point. The effect of suppressive interspike interactions on STDP is also analyzed and shown to modify the synaptic dynamics.

This contribution describes a neural network that self-organizes to recover the underlying original sources from typical sensor signals. No particular information is required about the statistical properties of the sources and the coefficients of the linear transformation, except the fact that the source signals are statistically independent and nonstationary. This is often true for real life applications. We propose an online learning solution using a neural network and use the nonstationarity of the sources to achieve the separation. The learning rule for the network's parameters is derived from the steepest descent minimization of a time-dependent cost function that takes the minimum only when the network outputs are uncorrelated with each other. In this process divide the problem into two learning problems one of which is solved by an anti-Hebbian learning and the other by an Hebbian learning process. We also compare the performance of our algorithm with other solutions to this task. r

The so far developed and widely utilized connectionist systems (artificial neural networks) are mainly based on a single brain-like connectionist principle of information processing, where learning and information exchange occur in the connections. This paper extends this paradigm of connectionist systems to a new trend-integrative connectionist learning systems (ICOS) that integrate in their structure and learning algorithms principles from different hierarchical levels of information processing in the brain, including neuronal-, genetic-, quantum. Spiking neural networks (SNN) are used as a basic connectionist learning model which is further extended with other information learning principles to create different ICOS. For example, evolving SNN for multitask learning are presented and illustrated on a case study of person authentification based on multimodal auditory and visual information. Integrative gene-SNN are presented, where gene interactions are included in the functioning of a spiking neuron. They are applied on a case study of computational neurogenetic modeling. Integrative quantum-SNN are introduced with a quantum Hebbian learning, where input features as well as information spikes are represented by quantum bits that result in exponentially faster feature selection and model learning. ICOS can be used to solve more efficiently challenging biological and engineering problems when fast adaptive learning systems are needed to incrementally learn in a large dimensional space. They can also help to better understand complex information processes in the brain especially how information processes at different information levels interact. Open questions, challenges and directions for further research are presented.

Todo lo que siempre quiso saber sobre Redes neuronales y nunca se atrevió a preguntar. Genealogía de los dispositivos conexionstas. El caso del perceptrón. Todas las clases de redes neuronales, excepto las redes generativas profundas creadas en la segunda década del siglo XXI a caballo de nuevas técnicas de aprendizaje profundo, protagonistas del nuevo boom de la Inteligencia Artificial.

This paper proposes a neural network that recovers some original random signals from their linear mixtures observed by the same number of sensors. The network acquires the function with a learning process without using any particular information about the statistical properties of the sources and the coefficients of the linear transformation, except the fact that the source signals are statistically independent and nonstationary. The learning rule for the network's parameters is derived from the steepest descent minimization of a time-dependent cost function that takes the minimum only when the network outputs are uncorrelated with each other.

In this paper we introduce two unsupervised techniques for visualization purposes based on the use of ensemble methods. The unsupervised techniques which are often quite sensitive to the presence of outliers are combined with the ensemble approaches in order to overcome the influence of outliers. The first technique is based on the use of Principal Component Analysis and the second

The principles by which spiking neurons contribute to the astounding computational power of generic cortical microcircuits, and how spike-timing-dependent plasticity (STDP) of synaptic weights could generate and maintain this computational function, are unknown. We show here that STDP, in conjunction with a stochastic soft winner-take-all (WTA) circuit, induces spiking neurons to generate through their synaptic weights implicit internal models for subclasses (or "causes") of the high-dimensional spike patterns of hundreds of pre-synaptic neurons. Hence these neurons will fire after learning whenever the current input best matches their internal model. The resulting computational function of soft WTA circuits, a common network motif of cortical microcircuits, could therefore be a drastic dimensionality reduction of information streams, together with the autonomous creation of internal models for the probability distributions of their input patterns. We show that the autonomous generation and maintenance of this computational function can be explained on the basis of rigorous mathematical principles. In particular, we show that STDP is able to approximate a stochastic online Expectation-Maximization (EM) algorithm for modeling the input data. A corresponding result is shown for Hebbian learning in artificial neural networks.