Multiscale stochastic approximation for parametric optimization of hidden Markov models

A two-time scale stochastic approximation algorithm is proposed for simulation-based parametric optimization of hidden Markov models, as an alternative to the traditional approaches to ''infinitesimal perturbation analysis.'' Its convergence is analyzed, and a queueing example is presented.

MULTI-INSTRUMENT DETECTION IN POLYPHONIC MUSIC USING GAUSSIAN MIXTURE BASED FACTORIAL HMM

We formulate the problem of detecting the constituent instruments in a polyphonic music piece as a joint decoding problem. From monophonic data, parametric Gaussian Mixture Hidden Markov Models (GM-HMM) are obtained for each instrument. We propose a method to use the above models in a factorial framework, termed as Factorial GM-HMM (F-GM-HMM). The states are jointly inferred to explain the evolution of each instrument in the mixture observation sequence. The dependencies are decoupled using variational inference technique. We show that the joint time evolution of all instruments' states can be captured using F-GM-HMM. We compare performance of proposed method with that of Student's-t mixture model (tMM) and GM-HMM in an existing latent variable framework. Experiments on two to five polyphony with 8 instrument models trained on the RWC dataset, tested on RWC and TRIOS datasets show that F-GM-HMM gives an advantage over the other considered models in segments containing co-occurring instruments.

On the DC Conductivity of HV DC Cable Insulation - Cautions in using the Empirical Models

The electrical conduction in insulating materials is a complex process and several theories have been suggested in the literature. Many phenomenological empirical models are in use in the DC cable literature. However, the impact of using different models for cable insulation has not been investigated until now, but for the claims of relative accuracy. The steady state electric field in the DC cable insulation is known to be a strong function of DC conductivity. The DC conductivity, in turn, is a complex function of electric field and temperature. As a result, under certain conditions, the stress at cable screen is higher than that at the conductor boundary. The paper presents detailed investigations on using different empirical conductivity models suggested in the literature for HV DC cable applications. It has been expressly shown that certain models give rise to erroneous results in electric field and temperature computations. It is pointed out that the use of these models in the design or evaluation of cables will lead to errors.

Discovery of Application Workloads from Network File Traces

An understanding of application I/O access patterns is useful in several situations. First, gaining insight into what applications are doing with their data at a semantic level helps in designing efficient storage systems. Second, it helps create benchmarks that mimic realistic application behavior closely. Third, it enables autonomic systems as the information obtained can be used to adapt the system in a closed loop.All these use cases require the ability to extract the application-level semantics of I/O operations. Methods such as modifying application code to associate I/O operations with semantic tags are intrusive. It is well known that network file system traces are an important source of information that can be obtained non-intrusively and analyzed either online or offline. These traces are a sequence of primitive file system operations and their parameters. Simple counting, statistical analysis or deterministic search techniques are inadequate for discovering application-level semantics in the general case, because of the inherent variation and noise in realistic traces.In this paper, we describe a trace analysis methodology based on Profile Hidden Markov Models. We show that the methodology has powerful discriminatory capabilities that enable it to recognize applications based on the patterns in the traces, and to mark out regions in a long trace that encapsulate sets of primitive operations that represent higher-level application actions. It is robust enough that it can work around discrepancies between training and target traces such as in length and interleaving with other operations. We demonstrate the feasibility of recognizing patterns based on a small sampling of the trace, enabling faster trace analysis. Preliminary experiments show that the method is capable of learning accurate profile models on live traces in an online setting. We present a detailed evaluation of this methodology in a UNIX environment using NFS traces of selected commonly used applications such as compilations as well as on industrial strength benchmarks such as TPC-C and Postmark, and discuss its capabilities and limitations in the context of the use cases mentioned above.

Identification of Active Sources in Single-Channel Convolutive Mixtures Using Known Source Models

We address the problem of identifying the constituent sources in a single-sensor mixture signal consisting of contributions from multiple simultaneously active sources. We propose a generic framework for mixture signal analysis based on a latent variable approach. The basic idea of the approach is to detect known sources represented as stochastic models, in a single-channel mixture signal without performing signal separation. A given mixture signal is modeled as a convex combination of known source models and the weights of the models are estimated using the mixture signal. We show experimentally that these weights indicate the presence/absence of the respective sources. The performance of the proposed approach is illustrated through mixture speech data in a reverberant enclosure. For the task of identifying the constituent speakers using data from a single microphone, the proposed approach is able to identify the dominant source with up to 8 simultaneously active background sources in a room with RT60 = 250 ms, using models obtained from clean speech data for a Source to Interference Ratio (SIR) greater than 2 dB.

Experiments and zero D modeling studies using specific Wiebe coefficients for producer gas as fuel in spark-ignited engines

The paper addresses experiments and modeling studies on the use of producer gas, a bio-derived low energy content fuel in a spark-ignited engine. Producer gas, generated in situ, has thermo-physical properties different from those of fossil fuel(s). Experiments on naturally aspirated and turbo-charged engine operation and subsequent analysis of the cylinder pressure traces reveal significant differences in the heat release pattern within the cylinder compared with a typical fossil fuel. The heat release patterns for gasoline and producer gas compare well in the initial 50% but beyond this, producer gas combustion tends to be sluggish leading to an overall increase in the combustion duration. This is rather unexpected considering that producer gas with nearly 20% hydrogen has higher flame speeds than gasoline. The influence of hydrogen on the initial flame kernel development period and the combustion duration and hence on the overall heat release pattern is addressed. The significant deviations in the heat release profiles between conventional fuels and producer gas necessitates the estimation of producer gas-specific Wiebe coefficients. The experimental heat release profiles are used for estimating the Wiebe coefficients. Experimental evidence of lower fuel conversion efficiency based on the chemical and thermal analysis of the engine exhaust gas is used to arrive at the Wiebe coefficients. The efficiency factor a is found to be 2.4 while the shape factor m is estimated at 0.7 for 2% to 90% burn duration. The standard Wiebe coefficients for conventional fuels and fuel-specific coefficients for producer gas are used in a zero D model to predict the performance of a 6-cylinder gas engine under naturally aspirated and turbo-charged conditions. While simulation results with standard Wiebe coefficients result in excessive deviations from the experimental results, excellent match is observed when producer gas-specific coefficients are used. Predictions using the same coefficients on a 3-cylinder gas engine having different geometry and compression ratio(s) indicate close match with the experimental traces highlighting the versatility of the coefficients.

Methodology for Pavement Design Reliability and Back Analysis Using Markov Chain Monte Carlo Simulation

Given the increasing cost of designing and building new highway pavements, reliability analysis has become vital to ensure that a given pavement performs as expected in the field. Recognizing the importance of failure analysis to safety, reliability, performance, and economy, back analysis has been employed in various engineering applications to evaluate the inherent uncertainties of the design and analysis. The probabilistic back analysis method formulated on Bayes' theorem and solved using the Markov chain Monte Carlo simulation method with a Metropolis-Hastings algorithm has proved to be highly efficient to address this issue. It is also quite flexible and is applicable to any type of prior information. In this paper, this method has been used to back-analyze the parameters that influence the pavement life and to consider the uncertainty of the mechanistic-empirical pavement design model. The load-induced pavement structural responses (e.g., stresses, strains, and deflections) used to predict the pavement life are estimated using the response surface methodology model developed based on the results of linear elastic analysis. The failure criteria adopted for the analysis were based on the factor of safety (FOS), and the study was carried out for different sample sizes and jumping distributions to estimate the most robust posterior statistics. From the posterior statistics of the case considered, it was observed that after approximately 150 million standard axle load repetitions, the mean values of the pavement properties decrease as expected, with a significant decrease in the values of the elastic moduli of the expected layers. An analysis of the posterior statistics indicated that the parameters that contribute significantly to the pavement failure were the moduli of the base and surface layer, which is consistent with the findings from other studies. After the back analysis, the base modulus parameters show a significant decrease of 15.8% and the surface layer modulus a decrease of 3.12% in the mean value. The usefulness of the back analysis methodology is further highlighted by estimating the design parameters for specified values of the factor of safety. The analysis revealed that for the pavement section considered, a reliability of 89% and 94% can be achieved by adopting FOS values of 1.5 and 2, respectively. The methodology proposed can therefore be effectively used to identify the parameters that are critical to pavement failure in the design of pavements for specified levels of reliability. DOI: 10.1061/(ASCE)TE.1943-5436.0000455. (C) 2013 American Society of Civil Engineers.

Joint evaluation of multiple speech patterns for speech recognition and training

We are addressing the novel problem of jointly evaluating multiple speech patterns for automatic speech recognition and training. We propose solutions based on both the non-parametric dynamic time warping (DTW) algorithm, and the parametric hidden Markov model (HMM). We show that a hybrid approach is quite effective for the application of noisy speech recognition. We extend the concept to HMM training wherein some patterns may be noisy or distorted. Utilizing the concept of ``virtual pattern'' developed for joint evaluation, we propose selective iterative training of HMMs. Evaluating these algorithms for burst/transient noisy speech and isolated word recognition, significant improvement in recognition accuracy is obtained using the new algorithms over those which do not utilize the joint evaluation strategy.

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.

An improved representation of vehicle incompatibility in frontal NCAP tests using a modified rigid barrier

With the objective of better understanding the significance of New Car Assessment Program (NCAP) tests conducted by the National Highway Traffic Safety Administration (NHTSA), head-on collisions between two identical cars of different sizes and between cars and a pickup truck are studied in the present paper using LS-DYNA models. Available finite element models of a compact car (Dodge Neon), midsize car (Dodge Intrepid), and pickup truck (Chevrolet C1500) are first improved and validated by comparing theanalysis-based vehicle deceleration pulses against corresponding NCAP crash test histories reported by NHTSA. In confirmation of prevalent perception, simulation-bascd results indicate that an NCAP test against a rigid barrier is a good representation of a collision between two similar cars approaching each other at a speed of 56.3 kmph (35 mph) both in terms of peak deceleration and intrusions. However, analyses carried out for collisions between two incompatible vehicles, such as an Intrepid or Neon against a C1500, point to the inability of the NCAP tests in representing the substantially higher intrusions in the front upper regions experienced by the cars, although peak decelerations in cars arc comparable to those observed in NCAP tests. In an attempt to improve the capability of a front NCAP test to better represent real-world crashes between incompatible vehicles, i.e., ones with contrasting ride height and lower body stiffness, two modified rigid barriers are studied. One of these barriers, which is of stepped geometry with a curved front face, leads to significantly improved correlation of intrusions in the upper regions of cars with respect to those yielded in the simulation of collisions between incompatible vehicles, together with the yielding of similar vehicle peak decelerations obtained in NCAP tests.

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.

Design of symmetric TIM barrel proteins from first principles

Background: Computational protein design is a rapidly maturing field within structural biology, with the goal of designing proteins with custom structures and functions. Such proteins could find widespread medical and industrial applications. Here, we have adapted algorithms from the Rosetta software suite to design much larger proteins, based on ideal geometric and topological criteria. Furthermore, we have developed techniques to incorporate symmetry into designed structures. For our first design attempt, we targeted the (alpha/beta)(8) TIM barrel scaffold. We gained novel insights into TIM barrel folding mechanisms from studying natural TIM barrel structures, and from analyzing previous TIM barrel design attempts. Methods: Computational protein design and analysis was performed using the Rosetta software suite and custom scripts. Genes encoding all designed proteins were synthesized and cloned on the pET20-b vector. Standard circular dichroism and gel chromatographic experiments were performed to determine protein biophysical characteristics. 1D NMR and 2D HSQC experiments were performed to determine protein structural characteristics. Results: Extensive protein design simulations coupled with ab initio modeling yielded several all-atom models of ideal, 4-fold symmetric TIM barrels. Four such models were experimentally characterized. The best designed structure (Symmetrin-1) contained a polar, histidine-rich pore, forming an extensive hydrogen bonding network. Symmetrin-1 was easily expressed and readily soluble. It showed circular dichroism spectra characteristic of well-folded alpha/beta proteins. Temperature melting experiments revealed cooperative and reversible unfolding, with a T-m of 44 degrees C and a Gibbs free energy of unfolding (Delta G degrees) of 8.0 kJ/mol. Urea denaturing experiments confirmed these observations, revealing a C-m of 1.6 M and a Delta G degrees of 8.3 kJ/mol. Symmetrin-1 adopted a monomeric conformation, with an apparent molecular weight of 32.12 kDa, and displayed well resolved 1D-NMR spectra. However, the HSQC spectrum revealed somewhat molten characteristics. Conclusions: Despite the detection of molten characteristics, the creation of a soluble, cooperatively folding protein represents an advancement over previous attempts at TIM barrel design. Strategies to further improve Symmetrin-1 are elaborated. Our techniques may be used to create other large, internally symmetric proteins.

Lesion Detection in Breast Ultrasound Images Using Tissue Transition Analysis

Breast cancer is one of the leading cause of cancer related deaths in women and early detection is crucial for reducing mortality rates. In this paper, we present a novel and fully automated approach based on tissue transition analysis for lesion detection in breast ultrasound images. Every candidate pixel is classified as belonging to the lesion boundary, lesion interior or normal tissue based on its descriptor value. The tissue transitions are modeled using a Markov chain to estimate the likelihood of a candidate lesion region. Experimental evaluation on a clinical dataset of 135 images show that the proposed approach can achieve high sensitivity (95 %) with modest (3) false positives per image. The approach achieves very similar results (94 % for 3 false positives) on a completely different clinical dataset of 159 images without retraining, highlighting the robustness of the approach.

ESR and optical studies of Mo5+ and W5+ ions in phosphomolybdate and phosphotungstate glasses

ESR and optical studies of phosphomolybdate and phosphotungstate glasses are discussed. Both the ESR and optical results indicate that molybdenum or tungsten ions are present in distorted octahedral environments in these glasses. In addition, ESR spectra of Mo5+ and W5+ ions show that the d electrons are localized on molybdenum and tungsten sites respectively. The variation of gperpendicular and gshort parallel values has been examined using appropriate structural models of these glasses.

Importance of non-conserved distal carboxyl terminal amino acids in two peptidases belonging to the M1 family: Thermoplasma acidophilum Tricorn interacting factor F2 and Escherichia coli Peptidase N

Enzymes belonging to the M1 family play important cellular roles and the key amino acids (aa) in the catalytic domain are conserved. However, C-terminal domain aa are highly variable and demonstrate distinct differences in organization. To address a functional role for the C-terminal domain, progressive deletions were generated in Tricorn interacting factor F2 from Thermoplasma acidophilum (F2) and Peptidase N from Escherichia coli (PepN). Catalytic activity was partially reduced in PepN lacking 4 C-terminal residues (PepNΔC4) whereas it was greatly reduced in F2 lacking 10 C-terminal residues (F2ΔC10) or PepN lacking eleven C-terminal residues (PepNΔC11). Notably, expression of PepNΔC4, but not PepNΔC11, in E. coliΔpepN increased its ability to resist nutritional and high temperature stress, demonstrating physiological significance. Purified C-terminal deleted proteins demonstrated greater sensitivity to trypsin and bound stronger to 8-amino 1-napthalene sulphonic acid (ANS), revealing greater numbers of surface exposed hydrophobic aa. Also, F2 or PepN containing large aa deletions in the C-termini, but not smaller deletions, were present in high amounts in the insoluble fraction of cell extracts probably due to reduced protein solubility. Modeling studies, using the crystal structure of E. coli PepN, demonstrated increase in hydrophobic surface area and change in accessibility of several aa from buried to exposed upon deletion of C-terminal aa. Together, these studies revealed that non-conserved distal C-terminal aa repress the surface exposure of apolar aa, enhance protein solubility, and catalytic activity in two soluble and distinct members of the M1 family.