The actor-critic algorithm as multi-time-scale stochastic approximation

The actor-critic algorithm of Barto and others for simulation-based optimization of Markov decision processes is cast as a two time Scale stochastic approximation. Convergence analysis, approximation issues and an example are studied.

Moving to stay in place: behavioral mechanisms for coexistence of African large carnivores

Most ecosystems have multiple predator species that not only compete for shared prey, but also pose direct threats to each other. These intraguild interactions are key drivers of carnivore community structure, with ecosystem-wide cascading effects. Yet, behavioral mechanisms for coexistence of multiple carnivore species remain poorly understood. The challenges of studying large, free-ranging carnivores have resulted in mainly coarse-scale examination of behavioral strategies without information about all interacting competitors. We overcame some of these challenges by examining the concurrent fine-scale movement decisions of almost all individuals of four large mammalian carnivore species in a closed terrestrial system. We found that the intensity of intraguild interactions did not follow a simple hierarchical allometric pattern, because spatial and behavioral tactics of subordinate species changed with threat and resource levels across seasons. Lions (Panthera leo) were generally unrestricted and anchored themselves in areas rich in not only their principal prey, but also, during periods of resource limitation (dry season), rich in the main prey for other carnivores. Because of this, the greatest cost (potential intraguild predation) for subordinate carnivores was spatially coupled with the highest potential benefit of resource acquisition (prey-rich areas), especially in the dry season. Leopard (P. pardus) and cheetah (Acinonyx jubatus) overlapped with the home range of lions but minimized their risk using fine-scaled avoidance behaviors and restricted resource acquisition tactics. The cost of intraguild competition was most apparent for cheetahs, especially during the wet season, as areas with energetically rewarding large prey (wildebeest) were avoided when they overlapped highly with the activity areas of lions. Contrary to expectation, the smallest species (African wild dog, Lycaon pictus) did not avoid only lions, but also used multiple tactics to minimize encountering all other competitors. Intraguild competition thus forced wild dogs into areas with the lowest resource availability year round. Coexistence of multiple carnivore species has typically been explained by dietary niche separation, but our multi-scaled movement results suggest that differences in resource acquisition may instead be a consequence of avoiding intraguild competition. We generate a more realistic representation of hierarchical behavioral interactions that may ultimately drive spatially explicit trophic structures of multi-predator communities.

STUDENT'S-t MIXTURE MODEL BASED MULTI-INSTRUMENT RECOGNITION IN POLYPHONIC MUSIC

We address the problem of multi-instrument recognition in polyphonic music signals. Individual instruments are modeled within a stochastic framework using Student's-t Mixture Models (tMMs). We impose a mixture of these instrument models on the polyphonic signal model. No a priori knowledge is assumed about the number of instruments in the polyphony. The mixture weights are estimated in a latent variable framework from the polyphonic data using an Expectation Maximization (EM) algorithm, derived for the proposed approach. The weights are shown to indicate instrument activity. The output of the algorithm is an Instrument Activity Graph (IAG), using which, it is possible to find out the instruments that are active at a given time. An average F-ratio of 0 : 7 5 is obtained for polyphonies containing 2-5 instruments, on a experimental test set of 8 instruments: clarinet, flute, guitar, harp, mandolin, piano, trombone and violin.

Toward a High-Resolution Monitoring of Continental Surface Water Extent and Dynamics, at Global Scale: from GIEMS (Global Inundation Extent from Multi-Satellites) to SWOT (Surface Water Ocean Topography)

Up to now, high-resolution mapping of surface water extent from satellites has only been available for a few regions, over limited time periods. The extension of the temporal and spatial coverage was difficult, due to the limitation of the remote sensing technique e.g., the interaction of the radiation with vegetation or cloud for visible observations or the temporal sampling with the synthetic aperture radar (SAR)]. The advantages and the limitations of the various satellite techniques are reviewed. The need to have a global and consistent estimate of the water surfaces over long time periods triggered the development of a multi-satellite methodology to obtain consistent surface water all over the globe, regardless of the environments. The Global Inundation Extent from Multi-satellites (GIEMS) combines the complementary strengths of satellite observations from the visible to the microwave, to produce a low-resolution monthly dataset () of surface water extent and dynamics. Downscaling algorithms are now developed and applied to GIEMS, using high-spatial-resolution information from visible, near-infrared, and synthetic aperture radar (SAR) satellite images, or from digital elevation models. Preliminary products are available down to 500-m spatial resolution. This work bridges the gaps and prepares for the future NASA/CNES Surface Water Ocean Topography (SWOT) mission to be launched in 2020. SWOT will delineate surface water extent estimates and their water storage with an unprecedented spatial resolution and accuracy, thanks to a SAR in an interferometry mode. When available, the SWOT data will be adopted to downscale GIEMS, to produce a long time series of water surfaces at global scale, consistent with the SWOT observations.

Recognition of Interaction Interface Residues in Low-Resolution Structures of Protein Assemblies Solely from the Positions of C alpha Atoms

Background: The number of available structures of large multi-protein assemblies is quite small. Such structures provide phenomenal insights on the organization, mechanism of formation and functional properties of the assembly. Hence detailed analysis of such structures is highly rewarding. However, the common problem in such analyses is the low resolution of these structures. In the recent times a number of attempts that combine low resolution cryo-EM data with higher resolution structures determined using X-ray analysis or NMR or generated using comparative modeling have been reported. Even in such attempts the best result one arrives at is the very course idea about the assembly structure in terms of trace of the C alpha atoms which are modeled with modest accuracy. Methodology/Principal Findings: In this paper first we present an objective approach to identify potentially solvent exposed and buried residues solely from the position of C alpha atoms and amino acid sequence using residue type-dependent thresholds for accessible surface areas of C alpha. We extend the method further to recognize potential protein-protein interface residues. Conclusion/Significance: Our approach to identify buried and exposed residues solely from the positions of C alpha atoms resulted in an accuracy of 84%, sensitivity of 83-89% and specificity of 67-94% while recognition of interfacial residues corresponded to an accuracy of 94%, sensitivity of 70-96% and specificity of 58-94%. Interestingly, detailed analysis of cases of mismatch between recognition of interface residues from C alpha positions and all-atom models suggested that, recognition of interfacial residues using C alpha atoms only correspond better with intuitive notion of what is an interfacial residue. Our method should be useful in the objective analysis of structures of protein assemblies when positions of only C alpha positions are available as, for example, in the cases of integration of cryo-EM data and high resolution structures of the components of the assembly.

A blind indic script recognizer for multi-script documents

We report a hierarchical blind script identifier for 11 different Indian scripts. An initial grouping of the 11 scripts is accomplished at the first level of this hierarchy. At the subsequent level, we recognize the script in each group. The various nodes of this tree use different feature-classifier combinations. A database of 20,000 words of different font styles and sizes is collected and used for each script. Effectiveness of Gabor and Discrete Cosine Transform features has been independently, evaluated using nearest neighbor linear discriminant and support vector machine classifiers. The minimum and maximum accuracies obtained, using this hierarchical mechanism, are 92.2% and 97.6%, respectively.

Joint decoding of multiple speech patterns for robust speech recognition

We are addressing a new problem of improving automatic speech recognition performance, given multiple utterances of patterns from the same class. We have formulated the problem of jointly decoding K multiple patterns given a single Hidden Markov Model. It is shown that such a solution is possible by aligning the K patterns using the proposed Multi Pattern Dynamic Time Warping algorithm followed by the Constrained Multi Pattern Viterbi Algorithm The new formulation is tested in the context of speaker independent isolated word recognition for both clean and noisy patterns. When 10 percent of speech is affected by a burst noise at -5 dB Signal to Noise Ratio (local), it is shown that joint decoding using only two noisy patterns reduces the noisy speech recognition error rate to about 51 percent, when compared to the single pattern decoding using the Viterbi Algorithm. In contrast a simple maximization of individual pattern likelihoods, provides only about 7 percent reduction in error rate.

Multi-Pattern Viterbi Algorithm for joint decoding of multiple speech patterns

Joint decoding of multiple speech patterns so as to improve speech recognition performance is important, especially in the presence of noise. In this paper, we propose a Multi-Pattern Viterbi algorithm (MPVA) to jointly decode and recognize multiple speech patterns for automatic speech recognition (ASR). The MPVA is a generalization of the Viterbi Algorithm to jointly decode multiple patterns given a Hidden Markov Model (HMM). Unlike the previously proposed two stage Constrained Multi-Pattern Viterbi Algorithm (CMPVA),the MPVA is a single stage algorithm. MPVA has the advantage that it cart be extended to connected word recognition (CWR) and continuous speech recognition (CSR) problems. MPVA is shown to provide better speech recognition performance than the earlier techniques: using only two repetitions of noisy speech patterns (-5 dB SNR, 10% burst noise), the word error rate using MPVA decreased by 28.5%, when compared to using individual decoding. (C) 2010 Elsevier B.V. All rights reserved.

Meridional propagation of large-scale monsoon convective zones

Observational studies indicate that the convective activity of the monsoon systems undergo intraseasonal variations with multi-week time scales. The zone of maximum monsoon convection exhibits substantial transient behavior with successive propagating from the North Indian Ocean to the heated continent. Over South Asia the zone achieves its maximum intensity. These propagations may extend over 3000 km in latitude and perhaps twice the distance in longitude and remain as coherent entities for periods greater than 2-3 weeks. Attempts to explain this phenomena using simple ocean-atmosphere models of the monsoon system had concluded that the interactive ground hydrology so modifies the total heating of the atmosphere that a steady state solution is not possible, thus promoting lateral propagation. That is, the ground hydrology forces the total heating of the atmosphere and the vertical velocity to be slightly out of phase, causing a migration of the convection towards the region of maximum heating. Whereas the lateral scale of the variations produced by the Webster (1983) model were essentially correct, they occurred at twice the frequency of the observed events and were formed near the coastal margin, rather than over the ocean. Webster's (1983) model used to pose the theories was deficient in a number of aspects. Particularly, both the ground moisture content and the thermal inertia of the model were severely underestimated. At the same time, the sea surface temperatures produced by the model between the equator and the model's land-sea boundary were far too cool. Both the atmosphere and the ocean model were modified to include a better hydrological cycle and ocean structure. The convective events produced by the modified model possessed the observed frequency and were generated well south of the coastline. The improved simulation of monsoon variability allowed the hydrological cycle feedback to be generalized. It was found that monsoon variability was constrained to lie within the bounds of a positive gradient of a convective intensity potential (I). The function depends primarily on the surface temperature, the availability of moisture and the stability of the lower atmosphere which varies very slowly on the time scale of months. The oscillations of the monsoon perturb the mean convective intensity potential causing local enhancements of the gradient. These perturbations are caused by the hydrological feedbacks, discussed above, or by the modification of the air-sea fluxes caused by variations of the low level wind during convective events. The final result is the slow northward propagation of convection within an even slower convective regime. The ECMWF analyses show very similar behavior of the convective intensity potential. Although it is considered premature to use the model to conduct simulations of the African monsoon system, the ECMWF analysis indicates similar behavior in the convective intensity potential suggesting, at least, that the same processes control the low frequency structure of the African monsoon. The implications of the hypotheses on numerical weather prediction of monsoon phenomenon are discussed.

Mapping of large scale 158 mu m [CII] line emission: Orion A

We present the first results of an observational programme undertaken to map the fine structure line emission of singly ionized carbon ([ CII] 157 : 7409 mum) over extended regions using a Fabry Perot spectrometer newly installed at the focal plane of a 100 cm balloon- borne far- infrared telescope. This new combination of instruments has a velocity resolution of similar to 200 km s(-1) and an angular resolution of 1.'5. During the first flight, an area of 30' x 15' in Orion A was mapped. These observations extend over a larger area than previous observations, the map is fully sampled and the spectral scanning method used enables reliable estimation of the continuum emission at frequencies adjacent to the [ CII] line. The total [ CII] line luminosity, calculated by considering up to 20% of the maximum line intensity is 0.04% of the luminosity of the far- infrared continuum. We have compared the [ CII] intensity distribution with the velocity- integrated intensity distributions of (CO)-C-13(1- 0), CI(1- 0) and CO( 3- 2) from the literature. Comparison of the [ CII], [ CI] and the radio continuum intensity distributions indicates that the largescale [ CII] emission originates mainly from the neutral gas, except at the position of M 43, where no [ CI] emission corresponding to the [ CII] emission is seen. Substantial part of the [ CII] emission from here originates from the ionized gas. The observed line intensities and ratios have been analyzed using the PDR models by Kaufman et al. ( 1999) to derive the incident UV flux and volume density at a few selected positions. The models reproduce the observations reasonably well at most positions excepting the [ CII] peak ( which coincides with the position of theta(1) Ori C). Possible reason for the failure could be the simplifying assumption of a homogeneous plane parallel slab in place of a more complicated geometry.

Comparison of AM-FM Based Features For Robust Speech Recognition

Effective feature extraction for robust speech recognition is a widely addressed topic and currently there is much effort to invoke non-stationary signal models instead of quasi-stationary signal models leading to standard features such as LPC or MFCC. Joint amplitude modulation and frequency modulation (AM-FM) is a classical non-parametric approach to non-stationary signal modeling and recently new feature sets for automatic speech recognition (ASR) have been derived based on a multi-band AM-FM representation of the signal. We consider several of these representations and compare their performances for robust speech recognition in noise, using the AURORA-2 database. We show that FEPSTRUM representation proposed is more effective than others. We also propose an improvement to FEPSTRUM based on the Teager energy operator (TEO) and show that it can selectively outperform even FEPSTRUM

Spatial Data Mining and Modeling for visualisation of Rapid Urbanisation

Rapid urbanisation in India has posed serious challenges to the decision makers in regional planning involving plethora of issues including provision of basic amenities (like electricity, water, sanitation, transport, etc.). Urban planning entails an understanding of landscape and urban dynamics with causal factors. Identifying, delineating and mapping landscapes on temporal scale provide an opportunity to monitor the changes, which is important for natural resource management and sustainable planning activities. Multi-source, multi-sensor, multi-temporal, multi-frequency or multi-polarization remote sensing data with efficient classification algorithms and pattern recognition techniques aid in capturing these dynamics. This paper analyses the landscape dynamics of Greater Bangalore by: (i) characterisation of direct impervious surface, (ii) computation of forest fragmentation indices and (iii) modeling to quantify and categorise urban changes. Linear unmixing is used for solving the mixed pixel problem of coarse resolution super spectral MODIS data for impervious surface characterisation. Fragmentation indices were used to classify forests – interior, perforated, edge, transitional, patch and undetermined. Based on this, urban growth model was developed to determine the type of urban growth – Infill, Expansion and Outlying growth. This helped in visualising urban growth poles and consequence of earlier policy decisions that can help in evolving strategies for effective land use policies.

Multi-Mode Phonon-Controlled Field Emission from Carbon Nanotubes: Modeling and Experiments

The main idea proposed in this paper is that in a vertically aligned array of short carbon nanotubes (CNTs) grown on a metal substrate, we consider a frequency dependent electric field, so that the mode-specific propagation of phonons, in correspondence with the strained band structure and the dispersion curves, take place. We perform theoretical calculations to validate this idea with a view of optimizing the field emission behavior of the CNT array. This is the first approach of its kind, and is in contrast to the the conventional approach where a DC bias voltage is applied in order to observe field emission. A first set of experimental results presented in this paper gives a clear indication that phonon-assisted control of field emission current in CNT based thin film diode is possible.

Integrated magnetics based multi-port bidirectional DC-DC converter topology for discontinuous-mode operation

A new type of multi-port isolated bidirectional DC-DC converter is proposed in this study. In the proposed converter, transfer of power takes place through addition of magnetomotive forces generated by multiple windings on a common transformer core. This eliminates the need for a centralised storage capacitor to interface all the ports. Hence, the requirement of an additional power transfer stage from the centralised capacitor can also be eliminated. The converter can be used for a multi-input, multi-output (MIMO) system. A pulse width modulation (PWM) strategy for controlling simultaneous power flow in the MIMO converter is also proposed. The proposed PWM scheme works in the discontinuous conduction mode. The leakage inductance can be chosen to aid power transfer. By using the proposed converter topology and PWM scheme, the need to compute power flow equations to determine the magnitude and direction of power flow between ports is alleviated. Instead, a simple controller structure based on average current control can be used to control the power flow. This study discusses the operating phases of the proposed multi-port converter along with its PWM scheme, the design process for each of the ports and finally experimental waveforms that validate the multi-port scheme.

High-temperature, high-pressure granulites (retrogressed eclogites) in the central region of the Lewisian, NW Scotland: Crustal-scale subduction in the Neoarchaean

Eclogites and associated high-pressure (HP) rocks in collisional and accretionary orogenic belts preserve a record of subduction and exhumation, and provide a key constraint on the tectonic evolution of the continents. Most eclogites that formed at high pressures but low temperatures at > 10-11 kbar and 450-650 degrees C can be interpreted as a result of subduction of cold oceanic lithosphere. A new class of high-temperature (HT) eclogites that formed above 900 degrees C and at 14 to 30 kbar occurs in the deep continental crust, but their geodynamic significance and processes of formation are poorly understood. Here we show that Neoarchaean mafic-ultramafic complexes in the central granulite facies region of the Lewisian in NW Scotland contain HP/HT garnet-bearing granulites (retrogressed eclogites), gabbros, Iherzolites, and websterites, and that the HP granulites have garnets that contain inclusions of omphacite. From thermodynamic modeling and compositional isopleths we calculate that peak eclogite-facies metamorphism took place at 24-22 kbar and 1060-1040 degrees C. The geochemical signature of one (G-21) of the samples shows a strong depletion of Eu indicating magma fractionation at a crustal level. The Sm-Nd isochron ages of HP phases record different cooling ages of ca. 2480 and 2330 Ma. We suggest that the layered mafic-ultramafic complexes, which may have formed in an oceanic environment, were subducted to eclogite depths, and exhumed as HP garnet-bearing orogenic peridotites. The layered complexes were engulfed by widespread orthogneisses of tonalite-trondhjemite-granodiorite (TTG) composition with granulite facies assemblages. We propose two possible tectonic models: (1) the fact that the relicts of eclogitic complexes are so widespread in the Scourian can be taken as evidence that a >90 km x 40 km-size slab of continental crust containing mafic-ultramafic complexes was subducted to at least 70 km depth in the late Archaean. During exhumation the gneiss protoliths were retrogressed to granulite facies assemblages, but the mafic-ultramafic rocks resisted retrogression. (2) The layered complexes of mafic and ultramafic rocks were subducted to eclogite-facies depths and during exhumation under crustal conditions they were intruded by the orthogneiss protoliths (TTG) that were metamorphosed in the granulite facies. Apart from poorly defined UHP metamorphic rocks in Norway, the retrogressed eclogites in the central granulite/retrogressed eclogite facies Lewisian region, NW Scotland have the highest crustal pressures so far reported for Archaean rocks, and demonstrate that lithospheric subduction was transporting crustal rocks to HP depths in the Neoarchaean. (C) 2012 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.