Exponentiated gradient algorithms for conditional random fields and max-margin Markov Networks

Log-linear and maximum-margin models are two commonly-used methods in supervised machine learning, and are frequently used in structured prediction problems. Efficient learning of parameters in these models is therefore an important problem, and becomes a key factor when learning from very large data sets. This paper describes exponentiated gradient (EG) algorithms for training such models, where EG updates are applied to the convex dual of either the log-linear or max-margin objective function; the dual in both the log-linear and max-margin cases corresponds to minimizing a convex function with simplex constraints. We study both batch and online variants of the algorithm, and provide rates of convergence for both cases. In the max-margin case, O(1/ε) EG updates are required to reach a given accuracy ε in the dual; in contrast, for log-linear models only O(log(1/ε)) updates are required. For both the max-margin and log-linear cases, our bounds suggest that the online EG algorithm requires a factor of n less computation to reach a desired accuracy than the batch EG algorithm, where n is the number of training examples. Our experiments confirm that the online algorithms are much faster than the batch algorithms in practice. We describe how the EG updates factor in a convenient way for structured prediction problems, allowing the algorithms to be efficiently applied to problems such as sequence learning or natural language parsing. We perform extensive evaluation of the algorithms, comparing them to L-BFGS and stochastic gradient descent for log-linear models, and to SVM-Struct for max-margin models. The algorithms are applied to a multi-class problem as well as to a more complex large-scale parsing task. In all these settings, the EG algorithms presented here outperform the other methods.

Mean and variance of estimates of the bispectrum of a harmonic random process-an analysis including leakage effects

Analytical expressions are derived for the mean and variance, of estimates of the bispectrum of a real-time series assuming a cosinusoidal model. The effects of spectral leakage, inherent in discrete Fourier transform operation when the modes present in the signal have a nonintegral number of wavelengths in the record, are included in the analysis. A single phase-coupled triad of modes can cause the bispectrum to have a nonzero mean value over the entire region of computation owing to leakage. The variance of bispectral estimates in the presence of leakage has contributions from individual modes and from triads of phase-coupled modes. Time-domain windowing reduces the leakage. The theoretical expressions for the mean and variance of bispectral estimates are derived in terms of a function dependent on an arbitrary symmetric time-domain window applied to the record. the number of data, and the statistics of the phase coupling among triads of modes. The theoretical results are verified by numerical simulations for simple test cases and applied to laboratory data to examine phase coupling in a hypothesis testing framework

Asphalt concrete surfaces macrotexture determination from still images

Road surface macrotexture is identified as one of the factors contributing to the surface's skid resistance. Existing methods of quantifying the surface macrotexture, such as the sand patch test and the laser profilometer test, are either expensive or intrusive, requiring traffic control. High-resolution cameras have made it possible to acquire good quality images from roads for the automated analysis of texture depth. In this paper, a granulometric method based on image processing is proposed to estimate road surface texture coarseness distribution from their edge profiles. More than 1300 images were acquired from two different sites, extending to a total of 2.96 km. The images were acquired using camera orientations of 60 and 90 degrees. The road surface is modeled as a texture of particles, and the size distribution of these particles is obtained from chord lengths across edge boundaries. The mean size from each distribution is compared with the sensor measured texture depth obtained using a laser profilometer. By tuning the edge detector parameters, a coefficient of determination of up to R2 = 0.94 between the proposed method and the laser profilometer method was obtained. The high correlation is also confirmed by robust calibration parameters that enable the method to be used for unseen data after the method has been calibrated over road surface data with similar surface characteristics and under similar imaging conditions.

Compilation of somatic mutations of the CDKN2 gene in human cancers : non-random distribution of base substitutions

The CDKN2 gene, encoding the cyclin-dependent kinase inhibitor p16, is a tumour suppressor gene that maps to chromosome band 9p21-p22. The most common mechanism of inactivation of this gene in human cancers is through homozygous deletion; however, in a smaller proportion of tumours and tumour cell lines intragenic mutations occur. In this study we have compiled a database of over 120 published point mutations in the CDKN2 gene from a wide variety of tumour types. A further 50 deletions, insertions, and splice mutations in CDKN2 have also been compiled. Furthermore, we have standardised the numbering of all mutations according to the full-length 156 amino acid form of p16. From this study we are able to define several hot spots, some of which occur at conserved residues within the ankyrin domains of p16. While many of the hotspots are shared by a number of cancers, the relative importance of each position varies, possibly reflecting the role of different carcinogens in the development of certain tumours. As reported previously, the mutational spectrum of CDKN2 in melanomas differs from that of internal malignancies and supports the involvement of UV in melanoma tumorigenesis. Notably, 52% of all substitutions in melanoma-derived samples occurred at just six nucleotide positions. Nonsense mutations comprise a comparatively high proportion of mutations present in the CDKN2 gene, and possible explanations for this are discussed.

Falls in Parkinson’s disease : evidence for altered stepping strategies on compliant surfaces

Background: Real-world environments comprise surfaces of different textures, densities and gradients, which can threaten postural stability and increase falls risk. However, there has been limited research that has examined how walking on compliant surfaces influences gait and postural stability in older people and PD patients. Methods: PD patients (n = 49) and age-matched controls (n = 32) were assessed using three dimensional motion analysis during self-paced walking on both firm and foam walkways. Falls were recorded prospectively over 12 months using daily falls calendars. Results: Walking on a foam surface influenced the temporospatial characteristics for all groups, but PD fallers adopted very different joint kinematics compared with controls. PD fallers also demonstrated reduced toe clearance and had increased mediolateral head motion(relative to walking velocity) compared with control participants. Conclusions: Postural control deficits in PD fallers may impair their capacity to attenuate surface-related perturbations and control head motion. The risk of falling for PD patients may be increased on less stable surfaces.

Driving Ge island ordering on nanostructured Si surfaces

Semiconductor epitaxial nanostructures have been recently proposed as the key building blocks of many innovative applications in materials science and technology. To bring their tremendous potential to fruition, a fine control of nanostructure size and placement is necessary. We present a detailed investigation of the self-ordering process in the prototype case of Ge/Si heteroepitaxy. Starting from a bottom-up strategy (step-bunching instabilities), our analysis moves to lithographic techniques (scanning tunneling lithography, nanomechanical stamping, focused ion beam patterning) with the aim of developing a hybrid approach in which the exogenous intervention is specifically designed to suit and harness the natural self-organization dynamics of the system.

A microanalytical study of the surfaces of normal, delipidized, and artificially “resurfaced” articular cartilage

The surface amorphous layer of articular cartilage is of primary importance to its load-bearing and lubrication function. This lipid-filled layer is degraded/disrupted or eliminated when cartilage degenerates due to diseases. This article examines further the characteristic of this surface overlay using a combination of microscopy and imaging methods to evaluate the hypothesis that the surface of articular cartilage can be repaired by exposing degraded cartilage to aqueous synthetic lipid mixtures. The preliminary results demonstrate that it is possible to create a new surface layer of phospholipids on the surface of cartilage following artificial lipid removal, but such a layer does not possess enough mechanical strength for physiological function when created with either unsaturated palmitoyloleoyl- phosphatidylcholine or saturated dipalmitoyl-phosphatidylcholine component of joint lipid composition alone. We conclude that this may be due to low structural cohesivity, inadequate time of exposure, and the mix/content of lipid in the incubation environment.