A global competitive neural network

 A study is presented of a set of coupled nets proposed to function as a global competitive network. One net, of hidden nodes, is composed solely of inhibitory neurons and is excitatorily driven and feeds back in a disinhibitory manner to an input net which itself feeds excitatorily to a (cortical) output net. The manner in which the former input and hidden inhibitory net function so as to enhance outputs as compared with inputs, and the further enhancements when the cortical net is added, are explored both mathematically and by simulation. This is extended to learning on cortical afferent and lateral connections. A global wave structure, arising on the inhibitory net in a similar manner to that of pattern formation in a negative laplacian net, is seen to be important to all of these activities. Simulations are only performed in one dimension, although the global nature of the activity is expected to extend to higher dimensions. Possible implications are briefly discussed. Received: 21 November 1993/Accepted in revised form: 30 June 1994     

A neural net model for epilepsy

A neural net model based in our previous studies with randomly interconnected neural nets is presented here capable of exhibiting epileptic features. These features can be explained in terms of the structural and dynamical properties of the model. In addition, apart from the fact that this model can imitate epileptic phenomena, it might also help to explain some poorly understood clinical phenomena from which general disturbances can produce focal seizures in the brain.     

A neural net model for masking phenomena

Some aspects of masking phenomena are considered in terms of the simplest possible model of two-factor neural elements. The effect of a number of variables can be accounted for, but the introduction of an internuncial element results in a masking function which need not be symmetric about zero delay interval. As an illustration, the results for a special case are compared with available data. In general, such a model results in a masking function which depends on the intensity, area, and duration of the stimuli, as well as on the temporal and spatial separation between them.     

A neural net model for escape learning

An escape learning situation is discussed in terms of a neural model in which a stimulus can result in a conditioned excitement and a specific conditioned response. By using the simplest relations between the strengths of conditioning and the number of reinforcements and by introducing a distribution of fluctuations occurring regularly in time, one can calculate the probabilities of various responses, as well as the various latencies, in successive trials. The results are in moderately satisfactory agreement with the data of R. L. Solomon and L. C. Wynne (Psychol. Monogr.,67, No. 4, 1953). Consequences of the model for various experimental situations are discussed. This research was supported in part by the United States Public Health Service Grant RCA GM K6 18,420 and in part by the United States Air Force through the Air Force Office of Scientific Research of the Air Research Development Command under Grant No. AF AFOSR 370-64.     

A neural net model for the α-rhythm

Pulse-interval distributions are obtained for a counting system in which there is a gradual, rather than abrupt, increase in excitability following the registration of a pulse (relative refractoriness). The results are applicable to systems in which Poisson counting would be observed in the absence of such effects, and in which the memory reaches back at most one pulse. Choosing a particular functional form for the recovery function, the theory fits the experimentally measured distribution for the maintained discharge in the cat's retinal ganglion cell. It is also consistent with the notion that Weber's Law emerges from refractoriness in the visual system, as first proposed by van der Velden.This work was supported in part by the National Science Foundation     

A neural net model for multiple memory domains

Previous studies with neural nets constructed of discrete populations of formal neurons have assumed that all neurons have the same probability of connection with any other neuron in the net. However, in this new study we incorporate the behavior of the neural systems in which the neural connections can be set up by means of chemical markers carried by the individual cells. With this new approach we studied the dynamics of isolated neural nets again as well as the dynamics of neural nets with sustained inputs. Results obtained with this approach show simple and multiple hysteresis phenomena. Such hysteresis loops may be considered to represent the basis for short-term memory.     

An avoidance learning situation. A neural net model

A simple avoidance situation is considered in terms of a neural net learning model. Data for the control situation can be represented by an expression having three parameters which determine the initial and the steady state activities together with the transient aspects. The introduction of a learning parameter then allows one to calculate satisfactorily the results obtained in the experimental situation in which shock is applied. This research was supported in part by the United States Air Force through the Air Force Office of Scientific Research of the Air Research Development Command under Grant No. AF AFOSR 370-63 and in part by the United States Public Health Service Grant RCA GM K6 18,420.     

A monoclonal antibody capable of modulating opioid binding to rat neural membranes

A monoclonal antibody capable of inhibiting opioid binding to rat neural membranes has been produced. Spleen cells from a BALB/c mouse, immunized with a partially purified opioid receptor complex, were fused with P3-X63.Ag8.653.3 myeloma cells. The cell line OR-689.2.4 secreted an IgM that was capable of partially inhibiting opioid binding to rat neural membranes under equilibrium binding conditions, while not affecting the binding of nonopioid ligands. Control mouse immunoglobulins and heat-denatured OR-689.2.4 did not inhibit opioid binding to membranes. The purified immunoglobulin inhibited the binding of [3H]dihydromorphine in a titrable, saturable, and reversible manner, as well as the binding of the delta-ligand [3H][D-Ala2,D-Leu5]enkephalin, the kappa-ligand [3H] ethylketocyclazocine, and 3H-labeled antagonists. In addition to blocking the binding of opioids to membranes, the immunoglobulin could also displace bound [3H]dihydromorphine from neural membranes. The 125I-labeled immunoglobulin specifically bound to neural membranes with a Kd of 1.3 nM and a maximal number of binding sites of 41.8 fmol/0.25 mg of membrane protein. In a titrable manner, the immunoglobulin precipitated opioid binding sites from a solubilized preparation of neural membranes. When OR-689.2.4 conjugated to Sepharose was incubated with the partially purified opioid receptor complex, labeled with 125I, a 35,000-dalton protein was specifically bound by the immunoglobulin. This antibody provides a tool for probing the multiple opioid binding sites.     

A biologically inspired neural net for trajectory formation and obstacle avoidance

In this paper we present a biologically inspired two-layered neural network for trajectory formation and obstacle avoidance. The two topographically ordered neural maps consist of analog neurons having continuous dynamics. The first layer, the sensory map, receives sensory information and builds up an activity pattern which contains the optimal solution (i.e. shortest path without collisions) for any given set of current position, target positions and obstacle positions. Targets and obstacles are allowed to move, in which case the activity pattern in the sensory map will change accordingly. The time evolution of the neural activity in the second layer, the motor map, results in a moving cluster of activity, which can be interpreted as a population vector. Through the feedforward connections between the two layers, input of the sensory map directs the movement of the cluster along the optimal path from the current position of the cluster to the target position. The smooth trajectory is the result of the intrinsic dynamics of the network only. No supervisor is required. The output of the motor map can be used for direct control of an autonomous system in a cluttered environment or for control of the actuators of a biological limb or robot manipulator. The system is able to reach a target even in the presence of an external perturbation. Computer simulations of a point robot and a multi-joint manipulator illustrate the theory.     

A note on the McCulloch-Pitts neural net for heat-cold discrimination

An error is pointed out in the neural net model which had been previously proposed to explain the heat-cold responses that depended on the duration of the stimulus. A corrected net for the effect is given. Contribution no. 3 of the Department of Biophysics of the University of Pittsburgh.     

Pattern computation in neural communication systems

Biological data suggests that activity patterns emerging in small- and large-scale neural systems may play an important role in performing the functions of the neural system, and in particular, neural computations. It is proposed in this paper that neural systems can be understood in terms of pattern computation and abstract communication systems theory. It is shown that analysing high-resolution surface EEG data, it is possible to determine abstract probabilistic rules that describe how emerging activity patterns follow earlier activity patterns. The results indicate the applicability of the proposed approach for understanding the working of complex neural systems.     

Normalization as a canonical neural computation

There is increasing evidence that the brain relies on a set of canonical neural computations, repeating them across brain regions and modalities to apply similar operations to different problems. A promising candidate for such a computation is normalization, in which the responses of neurons are divided by a common factor that typically includes the summed activity of a pool of neurons. Normalization was developed to explain responses in the primary visual cortex and is now thought to operate throughout the visual system, and in many other sensory modalities and brain regions. Normalization may underlie operations such as the representation of odours, the modulatory effects of visual attention, the encoding of value and the integration of multisensory information. Its presence in such a diversity of neural systems in multiple species, from invertebrates to mammals, suggests that it serves as a canonical neural computation.     

Oscillations in a refractory neural net

A functional differential equation that arises from the classic theory of neural networks is considered. As the length of the absolute refractory period is varied, there is, as shown here, a super-critical Hopf bifurcation. As the ratio of the refractory period to the time constant of the network increases, a novel relaxation oscillation occurs. Some approximations are made and the period of this oscillation is computed.     

Cyclic modes in neural net models

Neural network models with possible cross-coupling from every neuron to every other neuron are condisered. The all-or-none law is assumed for firing of neurons. A network is shown to have a set of latent cyclic modes. If the net is stimulated briefly, it will subsequently either return to quiescence or settle into periodic activity in one of its cyclic modes. Realization of cyclic modes and analysis of nets are discussed. A learning rule for adjustment of synaptic strengths is presented.     

FPNA: interaction between FPGA and neural computation

Neural networks are usually considered as naturally parallel computing models. But the number of operators and the complex connection graph of standard neural models can not be directly handled by digital hardware devices. More particularly, several works show that programmable digital hardware is a real opportunity for flexible hardware implementations of neural networks. And yet many area and topology problems arise when standard neural models are implemented onto programmable circuits such as FPGAs, so that the fast FPGA technology improvements can not be fully exploited. Therefore neural network hardware implementations need to reconcile simple hardware topologies with complex neural architectures. The theoretical and practical framework developed, allows this combination thanks to some principles of configurable hardware that are applied to neural computation: Field Programmable Neural Arrays (FPNA) lead to powerful neural architectures that are easy to map onto FPGAs, thanks to a simplified topology and an original data exchange scheme. This paper shows how FPGAs have led to the definition of the FPNA computation paradigm. Then it shows how FPNAs contribute to current and future FPGA-based neural implementations by solving the general problems that are raised by the implementation of complex neural networks onto FPGAs.     

A modular artificial neural net for controlling a six-legged walking system

 A system that controls the leg movement of an animal or a robot walking over irregular ground has to ensure stable support for the body and at the same time propel it forward. To do so, it has to react adaptively to unpredictable features of the environment. As part of our study of the underlying mechanisms, we present here a model for the control of the leg movement of a 6-legged walking system. The model is based on biological data obtained from the stick insect. It represents a combined treatment of realistic kinematics and biologically motivated, adaptive gait generation. The model extends a previous algorithmic model by substituting simple networks of artificial neurons for the algorithms previously used to control leg state and interleg coordination. Each system controlling an individual leg consists of three subnets. A hierarchically superior net contains two sensory and two ‘premotor’ units; it rhythmically suppresses the output of one or the other of the two subordinate nets. These are continuously active. They might be called the ‘swing module’ and the ‘stance module’ because they are responsible for controlling the swing (return stroke) and the stance (power stroke) movements, respectively. The swing module consists of three motor units and seven sensory units. It can produce appropriate return stroke movements for a broad range of initial and final positions, can cope with mechanical disturbances of the leg movement, and is able to react to an obstacle which hinders the normal performance of the swing movement. The complete model is able to walk at different speeds over irregular surfaces. The control system rapidly reestablishes a stable gait when the movement of the legs is disturbed. Received: 13 July 1994/Accepted in revised form: 15 November 1994     

A neural computation for visual acuity in the presence of eye movements

Humans can distinguish visual stimuli that differ by features the size of only a few photoreceptors. This is possible despite the incessant image motion due to fixational eye movements, which can be many times larger than the features to be distinguished. To perform well, the brain must identify the retinal firing patterns induced by the stimulus while discounting similar patterns caused by spontaneous retinal activity. This is a challenge since the trajectory of the eye movements, and consequently, the stimulus position, are unknown. We derive a decision rule for using retinal spike trains to discriminate between two stimuli, given that their retinal image moves with an unknown random walk trajectory. This algorithm dynamically estimates the probability of the stimulus at different retinal locations, and uses this to modulate the influence of retinal spikes acquired later. Applied to a simple orientation-discrimination task, the algorithm performance is consistent with human acuity, whereas naive strategies that neglect eye movements perform much worse. We then show how a simple, biologically plausible neural network could implement this algorithm using a local, activity-dependent gain and lateral interactions approximately matched to the statistics of eye movements. Finally, we discuss evidence that such a network could be operating in the primary visual cortex.     

Fly vision: neural mechanisms of motion computation

A wide range of novel approaches are being used to dissect the visual system of the fly, both the neural networks of motion detection and the performance of these networks under complex natural stimulus conditions.     

Statistical independence and neural computation in the leech ganglion

 In this report, the input/output relations in an isolated ganglion of the leech Hirudo medicinalis were studied by simultaneously using six or eight suction pipettes and two intracellular electrodes. Sensory input was mimicked by eliciting action potentials in mechanosensory neurons with intracellular electrodes. The integrated neural output was measured by recording extracellular voltage signals with pipettes sucking the roots and the connectives. A single evoked action potential activated electrical activity in at least a dozen different neurons, some of which were identified. This electrical activity was characterized by a high degree of temporal and spatial variability. The action potentials of coactivated neurons, i.e. activated by the same mechanosensory neuron, did not show any significant pairwise correlation. Indeed, the analysis of evoked action potentials indicates clear statistical independence among coactivated neurons, presumably originating from the independence of synaptic transmission at distinct synapses. This statistical independence may be used to increase reliability when neuronal activity is averaged or pooled. It is suggested that statistical independence among coactivated neurons may be a usual property of distributed processing of neuronal networks and a basic feature of neural computation. Received: 20 September 1999 / Accepted in revised form: 3 March 2000