Model selection and error estimation

We study model selection strategies based on penalized empirical loss minimization. We point out a tight relationship between error estimation and data-based complexity penalization: any good error estimate may be converted into a data-based penalty function and the performance of the estimate is governed by the quality of the error estimate. We consider several penalty functions, involving error estimates on independent test data, empirical {\sc vc} dimension, empirical {\sc vc} entropy, andmargin-based quantities. We also consider the maximal difference between the error on the first half of the training data and the second half, and the expected maximal discrepancy, a closely related capacity estimate that can be calculated by Monte Carlo integration. Maximal discrepancy penalty functions are appealing for pattern classification problems, since their computation is equivalent to empirical risk minimization over the training data with some labels flipped.

Random measurement error and regression dilution bias.

Summary points: - The bias introduced by random measurement error will be different depending on whether the error is in an exposure variable (risk factor) or outcome variable (disease) - Random measurement error in an exposure variable will bias the estimates of regression slope coefficients towards the null - Random measurement error in an outcome variable will instead increase the standard error of the estimates and widen the corresponding confidence intervals, making results less likely to be statistically significant - Increasing sample size will help minimise the impact of measurement error in an outcome variable but will only make estimates more precisely wrong when the error is in an exposure variable

Evaluation of heterogeneity of DNA ploidy in early gastric cancers.

DNA ploidy has been shown to be a predictive parameter for prognosis in various solid tumours. The prognostic value of DNA-ploidy in gastric cancers is still a matter of controversy. A possible explanation for the discrepant results reported in the literature could be sampling error in tumours with multiple stemlines differing in DNA-ploidy. In order to determine whether or not such heterogeneity exists in early gastric carcinoma, we have performed DNA cytophotometry on multiple samples of a group of 17 early gastric carcinomas, of which 8 were pure intramucosal and 9 were infiltrating into the submucosa. We found an aneuploid DNA-stemline in 8 (47%) early gastric cancers, more often in tumours invading into the submucosa (5/9) than in purely mucosal tumours (3/8). Multiple DNA-stemlines were found more frequently in submucosally infiltrating tumours (4/5). These results confirm the presence of DNA-aneuploid early gastric carcinoma which are frequently heterogeneous and suggest that heterogeneity occurs more frequently in tumours invading the submucosa. This heterogeneity is best detected by analysing multiple samples of tumours for DNA-ploidy.

Biomass and nutrient cycling in pure and mixed stands of native tree species in southeastern Bahia, Brazil

The objective of this paper is to study selected components of the nutrient cycle of pure and mixed stands of native forest species of Atlantic Forest in southeastern Brazil. Tree diameter, height, above-ground biomass, and nutrient content were determined in 22-year-old stands. Litterfall, litter decomposition, and nutrient concentration were evaluated from August 1994 to July 1995. The following species were studied: Peltogyne angustiflora, Centrolobium robustum, Arapatiella psilophylla, Sclerolobium chrysophyllum, Cordia trichotoma, Macrolobium latifolium. The litter of a natural forest and a 40-year-old naturally regenerated second-growth forest was sampled as well. The mixed-species outmatched pure stands in height, stem volume and total biomass (29.4 % more). The greatest amount of forest litter was observed in the natural forest (9.3 Mg ha-1), followed by the mixed-species stand (7.6 Mg ha-1) and secondary forest (7.3 Mg ha-1), and least litterfall was measured in the pure C. robustum stand (5.5 Mg ha-1). Litterfall seasonality varied among species in pure stands (CV from 44.7 to 91.4 %), unlike litterfall in the mixed-tree stand, where the variation was lower (CV 31.2 %). In the natural and second-growth forest, litterfall varied by 57.8 and 34.0 %, respectively. The annual rate of nutrient return via litterfall varied widely among forest ecosystems. Differences were detected between forest ecosystems in both the litter accumulation and quantity of litterlayer nutrients. The highest mean nutrient accumulation in above-ground biomass was observed in mixed-species stands. The total nutrient accumulation (N + P + K+ Ca + Mg) ranged from 0.97 to 1.93 kg tree-1 in pure stands, and from 1.21 to 2.63 kg tree-1 in mixed-species stands. Soil fertility under mixed-species stands (0-10 cm) was intermediate between the primary forest and pure-stand systems. The litterfall rate of native forest species in a mixed-species system is more constant, resulting in a more continuous decomposition rate. Consequently, both nutrient availability and quantity of organic matter in the soil are higher and the production system ecologically more sustainable.

Spatio-temporal Brain Dynamics Mediating Post-error Behavioral Adjustments.

Optimal behavior relies on flexible adaptation to environmental requirements, notably based on the detection of errors. The impact of error detection on subsequent behavior typically manifests as a slowing down of RTs following errors. Precisely how errors impact the processing of subsequent stimuli and in turn shape behavior remains unresolved. To address these questions, we used an auditory spatial go/no-go task where continual feedback informed participants of whether they were too slow. We contrasted auditory-evoked potentials to left-lateralized go and right no-go stimuli as a function of performance on the preceding go stimuli, generating a 2 × 2 design with "preceding performance" (fast hit [FH], slow hit [SH]) and stimulus type (go, no-go) as within-subject factors. SH trials yielded SH trials on the following trials more often than did FHs, supporting our assumption that SHs engaged effects similar to errors. Electrophysiologically, auditory-evoked potentials modulated topographically as a function of preceding performance 80-110 msec poststimulus onset and then as a function of stimulus type at 110-140 msec, indicative of changes in the underlying brain networks. Source estimations revealed a stronger activity of prefrontal regions to stimuli after successful than error trials, followed by a stronger response of parietal areas to the no-go than go stimuli. We interpret these results in terms of a shift from a fast automatic to a slow controlled form of inhibitory control induced by the detection of errors, manifesting during low-level integration of task-relevant features of subsequent stimuli, which in turn influences response speed.

Pure monoparesis: a particular stroke subgroup?

BACKGROUND: Acute stroke presenting as monoparesis is rare, with a pure motor deficit in the arm or leg extending to an isolated facial paresis. OBJECTIVE: To raise the question if acute stroke presenting as monoparesis is a different entity from stroke with a more extensive motor deficit. PATIENTS: In the Lausanne Stroke Registry (1979-2000), 195 (4.1%) of 4802 patients met the clinical criteria for pure monoparesis involving the face (22%), arm (63%), or leg (15%). RESULTS: In the vast majority of cases (> 95%), monoparesis corresponded to ischemic stroke with a favorable outcome, with initial computed tomography scans or magnetic resonance images showing no signs of hemorrhage. The lesion for a facial deficit was most frequently located subcortically (internal capsule); for an arm deficit, in the superficial middle cerebral artery; and for a leg deficit, in the anterior cerebral artery territory. In pure monoparesis, only 17% of the patients had more than 1 risk factor as compared with 26% of those with bimodal and trimodal hemiparesis and with 46% of all patients with stroke other than those with pure motor stroke. The only frequent risk factor was hypertension (53%); however, this frequency was no different from that in other patients with stroke. No major stroke etiology could be identified in any of the 3 subgroups of monoparesis. CONCLUSION: Our finding of a wide range of stroke localization and etiology in monoparesis without any particular subgroup suggests that no specific plan of investigation can be recommended for these patients.

A new estimator to minimize the error due to breathing in the measurement of respiratory impedance

This paper presents a new respiratory impedance estimator to minimize the error due to breathing. Its practical reliability was evaluated in a simulation using realistic signals. These signals were generated by superposing pressure and flow records obtained in two conditions: 1) when applying forced oscillation to a resistance- inertance- elastance (RIE) mechanical model; 2) when healthy subjects breathed through the unexcited forced oscillation generator. Impedances computed (4-32 Hz) from the simulated signals with the new estimator resulted in a mean value which was scarcely biased by the added breathing (errors less than 1 percent in the mean R, I , and E ) and had a small variability (coefficients of variation of R, I, and E of 1.3, 3.5, and 9.6 percent, respectively). Our results suggest that the proposed estimator reduces the error in measurement of respiratory impedance without appreciable extracomputational cost.

Local and global error models to improve uncertainty quantification

In groundwater applications, Monte Carlo methods are employed to model the uncertainty on geological parameters. However, their brute-force application becomes computationally prohibitive for highly detailed geological descriptions, complex physical processes, and a large number of realizations. The Distance Kernel Method (DKM) overcomes this issue by clustering the realizations in a multidimensional space based on the flow responses obtained by means of an approximate (computationally cheaper) model; then, the uncertainty is estimated from the exact responses that are computed only for one representative realization per cluster (the medoid). Usually, DKM is employed to decrease the size of the sample of realizations that are considered to estimate the uncertainty. We propose to use the information from the approximate responses for uncertainty quantification. The subset of exact solutions provided by DKM is then employed to construct an error model and correct the potential bias of the approximate model. Two error models are devised that both employ the difference between approximate and exact medoid solutions, but differ in the way medoid errors are interpolated to correct the whole set of realizations. The Local Error Model rests upon the clustering defined by DKM and can be seen as a natural way to account for intra-cluster variability; the Global Error Model employs a linear interpolation of all medoid errors regardless of the cluster to which the single realization belongs. These error models are evaluated for an idealized pollution problem in which the uncertainty of the breakthrough curve needs to be estimated. For this numerical test case, we demonstrate that the error models improve the uncertainty quantification provided by the DKM algorithm and are effective in correcting the bias of the estimate computed solely from the MsFV results. The framework presented here is not specific to the methods considered and can be applied to other combinations of approximate models and techniques to select a subset of realizations