Using active learning for monitoring networks design: The example of wind power plants sites evaluation

Locating new wind farms is of crucial importance for energy policies of the next decade. To select the new location, an accurate picture of the wind fields is necessary. However, characterizing wind fields is a difficult task, since the phenomenon is highly nonlinear and related to complex topographical features. In this paper, we propose both a nonparametric model to estimate wind speed at different time instants and a procedure to discover underrepresented topographic conditions, where new measuring stations could be added. Compared to space filling techniques, this last approach privileges optimization of the output space, thus locating new potential measuring sites through the uncertainty of the model itself.

Selective integration of auditory-visual looming cues by humans.

An object's motion relative to an observer can confer ethologically meaningful information. Approaching or looming stimuli can signal threats/collisions to be avoided or prey to be confronted, whereas receding stimuli can signal successful escape or failed pursuit. Using movement detection and subjective ratings, we investigated the multisensory integration of looming and receding auditory and visual information by humans. While prior research has demonstrated a perceptual bias for unisensory and more recently multisensory looming stimuli, none has investigated whether there is integration of looming signals between modalities. Our findings reveal selective integration of multisensory looming stimuli. Performance was significantly enhanced for looming stimuli over all other multisensory conditions. Contrasts with static multisensory conditions indicate that only multisensory looming stimuli resulted in facilitation beyond that induced by the sheer presence of auditory-visual stimuli. Controlling for variation in physical energy replicated the advantage for multisensory looming stimuli. Finally, only looming stimuli exhibited a negative linear relationship between enhancement indices for detection speed and for subjective ratings. Maximal detection speed was attained when motion perception was already robust under unisensory conditions. The preferential integration of multisensory looming stimuli highlights that complex ethologically salient stimuli likely require synergistic cooperation between existing principles of multisensory integration. A new conceptualization of the neurophysiologic mechanisms mediating real-world multisensory perceptions and action is therefore supported.

Energy balance, physical activity, and thermogenic effect of feeding in premature infants.

In order to assess the contribution of the thermogenic effect of feeding and muscular activity to total energy expenditure, nine premature infants were studied for 2 consecutive days during which time repeated measurements of energy expenditure by indirect calorimetry were performed throughout the day, combined with a visual activity score based on body movement. The infants were growing at 16.6 +/- 4.0 g/kg/day (mean +/- SD) and received 110 +/- 8 kcal/kg/day metabolizable energy (milk formula) and 522 +/- 40 mgN/kg/day. Their total energy expenditure was 68 +/- 4 kcal/kg/day indicating that 41 +/- 7 kcal/kg/day was retained for growth. Based on the combination of energy + N balances it was estimated that 80% of the weight gain was fat-free tissue and 20% was fat tissue. The rate of energy expenditure measured minute-by-minute was significantly and linearly correlated with the activity score in both the premeal (r = 0.75;p less than 0.001) and the postmeal periods (r = 0.74; p less than 0.001) with no difference in the regression slope, but with a significant difference in intercept. In preset feeding schedules the latter allowed an estimation of the thermogenic effect without the confounding effect of activity. This was found to be 3.1 +/- 1.8% when expressed as a percentage of metabolizable energy intake. However when the "classical" approach was used as a comparison (integration of extra energy expenditure induced by the meal), the thermogenic effect was found to be greater, i.e. 9.5 +/- 3.8% of the meal's metabolizable energy, due to the superimposed effect of physical activity in the postprandial state.(ABSTRACT TRUNCATED AT 250 WORDS)

A Cellular Perspective on Brain Energy Metabolism and Functional Imaging.

The energy demands of the brain are high: they account for at least 20% of the body's energy consumption. Evolutionary studies indicate that the emergence of higher cognitive functions in humans is associated with an increased glucose utilization and expression of energy metabolism genes. Functional brain imaging techniques such as fMRI and PET, which are widely used in human neuroscience studies, detect signals that monitor energy delivery and use in register with neuronal activity. Recent technological advances in metabolic studies with cellular resolution have afforded decisive insights into the understanding of the cellular and molecular bases of the coupling between neuronal activity and energy metabolism and point at a key role of neuron-astrocyte metabolic interactions. This article reviews some of the most salient features emerging from recent studies and aims at providing an integration of brain energy metabolism across resolution scales.