Optimality model of unsupervised spike-timing-dependent plasticity: synaptic memory and weight distribution

We studied the hypothesis that synaptic dynamics is controlled by three basic principles: (1) synapses adapt their weights so that neurons can effectively transmit information, (2) homeostatic processes stabilize the mean firing rate of the postsynaptic neuron, and (3) weak synapses adapt more slowly than strong ones, while maintenance of strong synapses is costly. Our results show that a synaptic update rule derived from these principles shares features, with spike-timing-dependent plasticity, is sensitive to correlations in the input and is useful for synaptic memory. Moreover, input selectivity (sharply tuned receptive fields) of postsynaptic neurons develops only if stimuli with strong features are presented. Sharply tuned neurons can coexist with unselective ones, and the distribution of synaptic weights can be unimodal or bimodal. The formulation of synaptic dynamics through an optimality criterion provides a simple graphical argument for the stability of synapses, necessary for synaptic memory.     

Optimal spike-timing-dependent plasticity for precise action potential firing in supervised learning

In timing-based neural codes, neurons have to emit action potentials at precise moments in time. We use a supervised learning paradigm to derive a synaptic update rule that optimizes by gradient ascent the likelihood of postsynaptic firing at one or several desired firing times. We find that the optimal strategy of up- and downregulating synaptic efficacies depends on the relative timing between presynaptic spike arrival and desired postsynaptic firing. If the presynaptic spike arrives before the desired postsynaptic spike timing, our optimal learning rule predicts that the synapse should become potentiated. The dependence of the potentiation on spike timing directly reflects the time course of an excitatory postsynaptic potential. However, our approach gives no unique reason for synaptic depression under reversed spike timing. In fact, the presence and amplitude of depression of synaptic efficacies for reversed spike timing depend on how constraints are implemented in the optimization problem. Two different constraints, control of postsynaptic rates and control of temporal locality, are studied. The relation of our results to spike-timing-dependent plasticity and reinforcement learning is discussed.     

Location-free haptic interaction for large-area social applications

In this paper, we discuss the potential of force perception technologies for realizing hand-held devices in the field of social systems. We propose and develop an interactive force-sensation-based navigation system for waiters based on a force perception technology that we have proposed. The navigation system consists of our new hand-held haptic interface and a camera-based position and posture identification system. Since the proposed compact haptic interface does not require external grounding, it can be used outside the laboratory and does not interrupt human activity. We verify the feasibility of the system in trials where we collected the responses of system users.     

GMDH utilizing BDS with insufficient modelling data

It has been pointed out that the group method of data handling algorithm has a disadvantage that predicts an abnormal result in the event of shortage of modelling data. This paper presents a method of modelling that avoids the emergence of abnormal predicted values. For estimation of partial polynomial parameters, a bad data suppression estimator is employed. It is based on a non-quadratic cost function that reduces to the weighted least squares estimator in the absence of bad data. Real data are used to illustrate that this method prevents abnormal results.     

Nearly extensive sequential memory lifetime achieved by coupled nonlinear neurons

Many cognitive processes rely on the ability of the brain to hold sequences of events in short-term memory. Recent studies have revealed that such memory can be read out from the transient dynamics of a network of neurons. However, the memory performance of such a network in buffering past information has been rigorously estimated only in networks of linear neurons. When signal gain is kept low, so that neurons operate primarily in the linear part of their response nonlinearity, the memory lifetime is bounded by the square root of the network size. In this work, I demonstrate that it is possible to achieve a memory lifetime almost proportional to the network size, "an extensive memory lifetime," when the nonlinearity of neurons is appropriately used. The analysis of neural activity revealed that nonlinear dynamics prevented the accumulation of noise by partially removing noise in each time step. With this error-correcting mechanism, I demonstrate that a memory lifetime of order [Formula: see text] can be achieved.     

A human-machine cooperative system for generating sign language animation using thermal image processing,a fuzzy algorithm,and simulated annealing

We propose a method for a sign language animation by skin region detection applied to an infrared thermal image. In a system incorporating the proposed method, a 3D CG model corresponding to a person’s characteristic posture while using sign language is generated automatically by pattern recognition of the thermal image, and then a person’s hand in the CG model is set. The hand part is made manually beforehand. If necessary, the model can be replaced manually by a more appropriate model corresponding to training key frames, and/or the same generated model can be refined manually. In our experiments, three hearing-impaired people, who were experienced in using sign language, recognized the Japanese sign language gestures of 70 words expressed as animations with 94.3% accuracy. We further improved the system by correcting the position and direction of the hand of the automatically generated model through the use of a fuzzy algorithm and simulated annealing.     

Electrostatic separation of plastics by friction mixer with rotary blades

A mixture of plastic pellets or waste plastic pieces is electrostatically separated. The mixture is charged by friction in a mixer and falls through a horizontal electric field of 4 kV/cm into three zones under the field. A mixture of two kinds of plastic pellets is separated with a purity of not less than 90%. A mixture of three kinds of pellets is separated to some extent in the zones at the extreme ends in the first separation, according to the locations of the different kinds of pellets in the triboelectric charging sequence. This mixture will be separated in successive separations following the first one into the respective pellet types. A mixer made of suitable plastic material and adapted to enhance the friction between different kinds of pellets assures easy and reliable separation. Proper mixing time, a high rotational speed of the mixer, and low humidity increase the purity of the separated pellets. Since the effect of blended additives of pellets on charging characteristics appears with good reproducibility, pellets and waste plastic pieces with such additives may likewise be successively separated into the respective plastics. © 1999 Scripta Technica, Electr Eng Jpn, 127(3): 33–40, 1999     

Interaction between coal and liquefaction catalysts studied by high temperature electron spin resonance

The high temperature electron spin resonance technique has been used to obtain in situ information on the behaviour of liquefaction catalysts during coal pyrolysis. The spin concentration in coal was induced in the presence of a catalyst at the pyrolysis temperature. ZnCl₂ drastically increased the spin concentration of coal. The order of activity of the catalysts with respect to the increase in spin concentration was: ZnCl₂ (impregnated) ?ZnCl₂ (dispersed) >ZnCl₂/KCl>SnCl₂ > SbCl₃≈AlCl₃ ≈CaCl₂ > coal alone.     

Dynamic Viscoelastic Behavior of F-actin on Heating

Although a relatively large amount of actin is contained in myofibril, the role of F-actin in the thermal gelation of muscle proteins has been little studied. We investigated the dynamic viscoelastic behavior of F-actin on heating to clarify the role of F-actin in the thermal gelation. F-actin did not form an elastic gel on heating but did turn into a curdy matter. In addition, F-actin showed no development either of the storage modulus or the loss modulus during a rise in temperature. These results suggest that F-actin negatively affects the gel formation of fish muscle proteins.     

Response characteristics of all-solid-state pH sensor using Li5YSi4O12 glass

A new type of all-solid-state pH sensor was investigated for the monitoring of pH in high temperature. The all-solid-state pH sensor consists of two half-cells: indicator electrode using the Li₅YSi₄O₁₂ glass and an Ag/AgCl reference electrode coated with Nafion film. A stable Nafion film was achieved by heat treating at 100 °C for 1 h. The electromotive force (EMF) of the all-solid-state pH sensor decreased linearly with pH increase in water in accordance with the Nernst's equation. The all-solid-state pH sensor operated stably up to 80 °C. The sensitivity of the all-solid-state pH sensor against pH was high, and the EMF was also scarcely influenced by the presence of inorganic ions such as Li⁺, Na⁺ and Cl⁻. It was practically confirmed by the pH titration test that the all-solid-state pH sensor behaved similar to the commercial pH meter with the conventional glass electrode. In addition, the all-solid-state pH sensor showed same equivalence point both at high temperature and low temperature operations.