Trial and Error

A pac-learning algorithm is -space bounded, if it stores at most examples from the sample at any time. We characterize the -space learnable concept classes. For this purpose we introduce the compression parameter of a concept class and design our Trial and Error Learning Algorithm. We show : is -space learnable if and only if the compression parameter of is at most . This learning algorithm does not produce a hypothesis consistent with the whole sample as previous approaches e.g. by Floyd, who presents consistent space bounded learning algorithms, but has to restrict herself to very special concept classes. On the other hand our algorithm needs large samples; the compression parameter appears as exponent in the sample size. We present several examples of polynomial time space bounded learnable concept classes: – all intersection closed concept classes with finite VC–dimension. – convex -gons in . – halfspaces in . – unions of triangles in . We further relate the compression parameter to the VC–dimension, and discuss variants of this parameter. Received May 24, 1994 / July 4, 1995     

Comportement anodique de Ga(As1-txPx) sous illumination

The anodic behaviour of Ga(As_1-xP_x) in basic media and under illumination has been investigated by measuring the electrode characteristics and the interfacial capacity values. These measurements are interpreted using Gärtner's classical model. The diffusion length of minority carriers and the gap value have been calculated. It is necessary to assume a recombination on the surface to explain the variation to the photolyse current density with the potential. This process may be attributed to a centre for which it has been possible to evaluate the energy level.     

The Price and Trading Volume Dynamics Relationship in the EEX Power Market: A Wavelet Modeling

This paper examines the dynamic relationship between power spot prices and related trading volumes in one of the most emergent energy markets. Traditionally, investigating the bivariate stochastic processes has been dominated by linear econometrical methods that proved helpful especially in finance. However, when dealing with intradaily power data, we cannot rely on models developed for financial or other commodity markets. Therefore, wavelet transforms are applied for power markets data to search for and decode nonlinear regularities and hidden patterns existing between the variables. Given its ability to decompose the time series into their time scale components and thus to reveal structure at different time horizons, wavelets are useful in analyzing situations in which the degree of association between processes is likely to change with the time-horizon. Therefore, a wavelet-based cross-analysis is performed between prices and trading volume time series. On the same basis, causality tests and out-of-sample forecasting tasks are carried out to empirically the strong relationship between the two investigated time series.     

Clinical Determination of Dry Body Weight

While nephrologists wait for the ideal, non invasive, inexpensive, precise, and reproducible tool to evaluate extracellular volume (ECV), they need to exert their clinical acumen in the quest of that holy grail, dry weight (DW). Estimation of DW using a clinical approach based on blood pressure (BP) and ECV is feasible and reliable as shown by successful experiences in various dialysis modes over more than three decades. But a need still exists to resolve difficulties associated with accurate assessment of BP (methods and circumstances of measurement, and the confounding effects of antihypertensive drugs) and ECV (evaluation of weight changes unrelated to ECV, lack of specificity and sensitivity of clinical symptoms, lag time, confusion in terminology). An essential point in clinical assessment of DW is that a normal BP is at the same time the target and the crucial index of DW achievement. For this reason, a trialand‐error “probe” process has to be used at intervals to make sure that the dry weight target point is correctly estimated. The various “non clinical” methods proposed for dry weight assessment increase the complexity and the cost of hemodialysis. They are, in the present state of things, more clinical research than practice tools. They do not replace clinical judgment.     

Sintering and mechanical properties of tricalcium phosphate–fluorapatite composites

Tricalcium phosphate and synthesized fluorapatite powder were mixed in order to elaborate biphasic ceramics composites. The effect of fluorapatite addition on the densification and the mechanical properties of tricalcium phosphate were measured with the change in composition and microstructure of the bioceramic. The Brazilian test was used to measure the mechanical resistance of the tricalcium phosphate–26.52 wt% fluorapatite composites. The densification and rupture strength increase versus sintering temperature. The composites have a good sinterability and rupture strength in temperature ranging between 1300 and 1400 °C. Thus, the densification ultimate was obtained at 1350 °C and the mechanical resistance optimum reached 9.6 MPa at 1400 °C. Above 1400 °C, the densification and the mechanical properties were hindered by the allotropic transformation of tricalcium phosphate, grain growth and the formation of both intragranular porosity and many cracks. The ³¹P magic angle spinning nuclear magnetic resonance analysis of composites reveals the presence of tetrahedral P sites.     

Trial and error

A pac-learning algorithm isd-space bounded, if it stores at mostd examples from the sample at any time. We characterize thed-space learnable concept classes. For this purpose we introduce the compression parameter of a concept classb and design our Trial and Error Learning Algorithm. We show: b isd-space learnable if and only if the compression parameter ofb is at mostd. This learning algorithm does not produce a hypothesis consistent with the whole sample as previous approaches e.g. by Floyd, who presents consistent space bounded learning algorithms, but has to restrict herself to very special concept classes. On the other hand our algorithm needs large samples; the compression parameter appears as exponent in the sample size. We present several examples of polynomial time space bounded learnable concept classes:

- all intersection closed concept classes with finite VC-dimension.

- convexn-gons in ?².

- halfspaces in ?ⁿ.

- unions of triangles in ?².

We further relate the compression parameter to the VC-dimension, and discuss variants of this parameter.     

Finite-time stabilization of uncertain delayed-hopfield neural networks with a time-varying leakage delay via non-chattering control

This article is concerned with the finite-time stabilization(FTSB) of a class of delayed-Hopfield neural networks with a timevarying delay in the leakage term in the presence of parameter uncertainties. To accomplish the target of FTSB, two new finitetime controllers are designed for uncertain delayed-Hopfield neural networks with a time-varying delay in the leakage term. By utilizing the finite-time stability theory and the Lyapunov-Krasovskii functional(LKF) approach, some sufficient conditions for the FTSB of these neural networks are established. These conditions, which can be used for the selection of control parameters,are in the form of linear matrix inequalities(LMIs) and can be numerically checked. Additionally, an upper bound of the settling time was estimated. Finally, our theoretical results are further substantiated by two numerical examples with graphical illustrations to demonstrate the effectiveness of the results.     

Heat transfer coefficient and thermal losses of solar collector and Nusselt number correlation for rectangular solar air heater duct with longitudinal fins hold under the absorber plate

An experimental investigation has been carried out for a series of system and operating parameters in order to analyze the effect of mass flow rate on heat transfer and Nusselt number characteristics in solar air heater. Experiments are performed at different air mass flow rates; varying from 0.012 to 0.016 kg/s, about hot summer days of Mai 2012. Hourly values of global solar radiation and some meteorological data (temperature, wind speed, relative humidities, etc.) for measuring days are obtained from the Biskra city of Algeria. The experiments encompassed the flow Reynolds number in the range 965.48–1301.4. Longitudinal fins were used inferior the absorber plate for an increase the heat exchange and render the flow fluid in the channel uniform. The effects of mass flow rate of air on the outlet temperature, Nusselt Number, Reynolds Number, Prandtl Number, the heat transfer in the thickness and length of the solar air collector were studied. For this effect was have created a new correlation correspondent of solar air collector with using fins it was written Nu = κ₀Re^1.36Pr^?0.68exp(0.342m)h ^[?0.018Pr].     

Changes in EEG activity before and after exhaustive exercise in sedentary women in neutral and hot environments

This study examined the effect of hyperthermia on brain electrical activity measured with encephalography during prolonged exhaustive exercise in a group of sedentary women (VO₂max = 35 ± 4 mL kg min⁻¹). Two strenuous cycling exercises were performed either in neutral (N-Ex) or in heat (H-Ex) conditions. Tympanic temperature (Tty), heart rate (HR), body mass loss (BML), plasma volume decrease, and brain electrical activity [EEG: α (8–13 Hz) and β(13–30 Hz)-band and α/β index of fatigue: the ratio between EEG activity in the α band and β-band] were recorded throughout the cycling sessions. The Tty increase 1.0 °C in the N-Ex and 1.8 °C in H-Ex. HR increased in both sessions but with significantly higher values during the H-Ex session when compared with the N-Ex session (p < 0.001) (from 85 ± 4 beats min⁻¹ to 164 ± 6 beats min⁻¹ and from 83 ± 6 beats min⁻¹ to 181 ± 8 beats min⁻¹, respectively in N-Ex and in H-Ex). This was associated with a significantly higher BML (p < 0.05) and a higher plasma volume decrease in the H-Ex session (p < 0.01). The α/β index increased significantly during both trials particularly during the H-Ex session (p < 0.05). This was associated with a significant decrease of time to exhaustion (−34%). We suggest that exhausting work in the heat induced a change in gross brain activity (alpha/beta ratio) compared to a longer, less thermally demanding exposure. Fatigue in the heat could be attributed to central factors as well as thermal, cardiac and hydro-electrolytic impairment.