Efficient distribution-free population learning of simple concepts

We consider a variant of the ‘population learning model’ proposed by Kearns and Seung [8], in which the learner is required to be ‘distribution-free’ as well as computationally efficient. A population learner receives as input hypotheses from a large population of agents and produces as output its final hypothesis. Each agent is assumed to independently obtain labeled sample for the target concept and output a hypothesis. A polynomial time population learner is said to PAC-learn a concept class, if its hypothesis is probably approximately correct whenever the population size exceeds a certain bound which is polynomial, even if the sample size for each agent is fixed at some constant. We exhibit some general population learning strategies, and some simple concept classes that can be learned by them. These strategies include the ‘supremum hypothesis finder’, the ‘minimum superset finder’ (a special case of the ‘supremum hypothesis finder’), and various voting schemes. When coupled with appropriate agent algorithms, these strategies can learn a variety of simple concept classes, such as the ‘high–low game’, conjunctions, axis-parallel rectangles and others. We give upper bounds on the required population size for each of these cases, and show that these systems can be used to obtain a speed up from the ordinary PAC-learning model [11], with appropriate choices of sample and population sizes. With the population learner restricted to be a voting scheme, what we have is effectively a model of ‘population prediction’, in which the learner is to predict the value of the target concept at an arbitrarily drawn point, as a threshold function of the predictions made by its agents on the same point. We show that the population learning model is strictly more powerful than the population prediction model. Finally, we consider a variant of this model with classification noise, and exhibit a population learner for the class of conjunctions in this model. This revised version was published online in June 2006 with corrections to the Cover Date.     

Connection‐Strength Estimation of Neuronal Networks by Fitting for Izhikevich Model

Recently, there has been abundant research using multineuron recording, but there are many problems with extracting the features from the obtained spike time series, which are huge in volume and complex. Here we introduce a new method of estimating synaptic connection strengths between neurons by fitting to the Izhikevich model by maximum likelihood estimation. We demonstrate that our method can estimate connection strengths from spike time series given by a simulated neural ensemble and can estimate nonconnectivity between two independent cultured neuronal networks. These results suggest that our method is applicable to network and plasticity analysis of neuronal networks.     

A FAST METHOD OF RADIATIVE HEAT TRANSFER ANALYSIS BETWEEN ARBITRARY THREE-DIMENSIONAL BODIES COMPOSED OF SPECULAR AND DIFFUSE SURFACES

The radiation element method by ray emission model (REM²) has been improved by using the law of reciprocity for the specular view factor and the incomplete Cholesky conjugate gradient (ICCG) method to reduce computational time. This improved method was applied to analyze the radiative heat transfer between arbitrary three-dimensional bodies composed of specular and diffuse surfaces. The accuracy of the improved method was evaluated by comparing analytical solutions. And the method was used to calculate radiative heat transfer between machine elements and heater panels. CPU time to calculate the radiative exchange for a model composed of 3193 elements was reduced to 1/430 of that by the previous numerical method using a decomposition method.     

2-Arachidonoyl glycerol suppresses gastric emptying via the cannabinoid receptor 1-cholecystokinin signaling pathway in mice

2-Monoacylglycerol (2-MAG) is one of the digestion products of dietary lipids. We recently demonstrated that a 2-MAG, 2-arachidonoyl glycerol (2-AG) potently stimulated cholecystokinin (CCK) secretion via cannabinoid receptor 1 (CB1) in a murine CCK-producing cell line, STC-1. CCK plays a crucial role in suppressing postprandial gastric emptying. To examine the effect of 2-AG on gastric emptying, we performed acetaminophen and phenol red recovery tests under oral or intraperitoneal administration of 2-AG in mice. Orally administered 2-AG (25 mg/kg) suppressed the gastric emptying rate in mice, as determined by the acetaminophen absorption test and phenol red recovery test. Intraperitoneal administration of a cholecystokinin A receptor antagonist (0.5 mg/kg) attenuated the gastric inhibitory emptying effect. In addition, both oral (10 mg/kg) and intraperitoneal (0.5 mg/kg) administration of a CB1 antagonist counteracted the 2-AG-induced gastric inhibitory effect. Furthermore, intraperitoneal 2-AG (25 mg/kg) suppressed gastric emptying. These results indicate that 2-AG exhibits an inhibitory effect on gastric emptying in mice, possibly mediated by stimulating both CCK secretion via CB1 expressed in CCK-producing cells and acting on CB1 expressed in the peripheral nerves. Our findings provide novel insights into the 2-MAG-sensing mechanism in enteroendocrine cells and the physiological role of 2-MAG.     

Composition of Corn Hull Dietary Fiber

The corn hull dietary fiber was decomposed into their components such as cellulose, hemicellulose-A, -B, -C, -other and lignin by the Siegel method, and their contents were 16%, 1%, 57%, 14% and 2%, respectivity. The components of ordinary corn fiber were insoluble in water, but several decomposed fractions were able to dissolved. The solubility in water of hemicellulose-B was well and of hemicellulose A was slightly, but other fractions were insoluble. Hemicellulose-A, -B, -C and -other fractions were made up of about 30% of arabinose and 50% of xylose. Uronic acid contents and solubility in water of each hemicellulose fraction were mutually related.