Predictability of music descriptor time series and its application to cover song detection

Intuitively, music has both predictable and unpredictable components. In this work we assess this qualitative statement in a quantitative way using common time series models fitted to state-of-the-art music descriptors. These descriptors cover different musical facets and are extracted from a large collection of real audio recordings comprising a variety of musical genres. Our findings show that music descriptor time series exhibit a certain predictability not only for short time intervals, but also for mid-term and relatively long intervals. This fact is observed independently of the descriptor, musical facet and time series model we consider. Moreover, we show that our findings are not only of theoretical relevance but can also have practical impact. To this end we demonstrate that music predictability at relatively long time intervals can be exploited in a real-world application, namely the automatic identification of cover songs (i.e. different renditions or versions of the same musical piece). Importantly, this prediction strategy yields a parameter-free approach for cover song identification that is substantially faster, allows for reduced computational storage and still maintains highly competitive accuracies when compared to state-of-the-art systems.

Parallel Research, Multiple Intellectual Property Right Protection Instruments, and the Correlation among R&D Projects, September 2005

The choice of a research path in attacking scientific and technological problems is a significant component of firms’ R&D strategy. One of the findings of the patent races literature is that, in a competitive market setting, firms’ noncooperative choices of research projects display an excessive degree of correlation, as compared to the socially optimal level. The paper revisits this question in a context in which firms have access to trade secrets, in addition to patents, to assert intellectual property rights (IPR) over their discoveries. We find that the availability of multiple IPR protection instruments can move the paths chosen by firms engaged in an R&D race toward the social optimum.

Distribución factorial del ingreso en América Latina, 1950-2000: nuevas series a partir de las cuentas nacionales

Este artículo presenta nuevas series de distribución factorial del ingreso entre 1950 y 2000 para 14 países de América Latina a partir de un proceso de armonización de los datos de compensación de los trabajadores recogidos en las cuentas nacionales. Además, se presentan estimaciones de la remuneración a los trabajadores autónomos, discutiéndose las diferentes metodologías de estimación posibles, y las limitaciones de datos que afectan a las mismas. El análisis de estas estimaciones coherentes nos permite llegar a algunas conclusionespreliminares. En primer lugar, las estimaciones presentan variaciones importantes entre los países. Segundo, a nivel regional, muestran variaciones tanto coyunturales como de largo plazo que apoyanlos estudios que cuestionan la estabilidad de la distribución factorial del ingreso en el largo plazo.Tercero, nuestras estimaciones de la remuneración del factor trabajo, una vez corregidas para incluir una estimación del trabajo no asalariado, siguen siendo sensiblemente inferiores a las de lospaíses desarrollados, cuestionando así los estudios que señalan que dichas variaciones desaparecen al aplicárseles esta corrección.

La construcción del espacio dramático en la ficción televisiva norteamericana. Análisis del espacio vivencial en las series dramáticas de la HBO

El presente trabajo analiza la construcción del espacio vivencial en cinco series televisivas producidas por la cadena norteamericana HBO: Carnivàle, Deadwood, The Sopranos, The Wire y Treme. La investigación parte de la convicción de una posible reelaboración de la historia del espacio doméstico norteamericano a través de la puesta en escena y de la potencialidad simbólica del espacio serial. De este modo, el trabajo propone un recorrido hermenéutico en el que se analiza la evolución del espacio narrativo, su evolución morfológica en dependencia de las necesidades diegéticas de cada una de las obras analizadas.

Cyclical Skill-Biased Technological Change

Over the past two decades, technological progress has been biased towards making skilled labor more productive. The evidence for this finding is based on the persistent parallel increase in the skill premium and the supply of skilled workers. What are the implications of skill-biased technological change for the business cycle? To answer this question, we use the CPS outgoing rotation groups to construct quarterly series for the price and quantity of skill. The unconditional correlation of the skill premium with the cycle is zero. However, using a structural VAR with long run restrictions, we find that technology shocks substantially increase the premium. Investment-specific technology shocks are not skill-biased and our findings suggest that capital and skill are (mildly) substitutable in aggregate production.

Time Series Input Selection using Multiple Kernel Learning

In this paper we study the relevance of multiple kernel learning (MKL) for the automatic selection of time series inputs. Recently, MKL has gained great attention in the machine learning community due to its flexibility in modelling complex patterns and performing feature selection. In general, MKL constructs the kernel as a weighted linear combination of basis kernels, exploiting different sources of information. An efficient algorithm wrapping a Support Vector Regression model for optimizing the MKL weights, named SimpleMKL, is used for the analysis. In this sense, MKL performs feature selection by discarding inputs/kernels with low or null weights. The approach proposed is tested with simulated linear and nonlinear time series (AutoRegressive, Henon and Lorenz series).

Parallel imaging with phase scrambling.

PURPOSE: Most existing methods for accelerated parallel imaging in MRI require additional data, which are used to derive information about the sensitivity profile of each radiofrequency (RF) channel. In this work, a method is presented to avoid the acquisition of separate coil calibration data for accelerated Cartesian trajectories. METHODS: Quadratic phase is imparted to the image to spread the signals in k-space (aka phase scrambling). By rewriting the Fourier transform as a convolution operation, a window can be introduced to the convolved chirp function, allowing a low-resolution image to be reconstructed from phase-scrambled data without prominent aliasing. This image (for each RF channel) can be used to derive coil sensitivities to drive existing parallel imaging techniques. As a proof of concept, the quadratic phase was applied by introducing an offset to the x(2) - y(2) shim and the data were reconstructed using adapted versions of the image space-based sensitivity encoding and GeneRalized Autocalibrating Partially Parallel Acquisitions algorithms. RESULTS: The method is demonstrated in a phantom (1 × 2, 1 × 3, and 2 × 2 acceleration) and in vivo (2 × 2 acceleration) using a 3D gradient echo acquisition. CONCLUSION: Phase scrambling can be used to perform parallel imaging acceleration without acquisition of separate coil calibration data, demonstrated here for a 3D-Cartesian trajectory. Further research is required to prove the applicability to other 2D and 3D sampling schemes. Magn Reson Med, 2014. © 2014 Wiley Periodicals, Inc.

A la recherche de symptômes de désorientation dépendant d'un déficit d'intégration multisensorielle chez l'animal : un éclairage sur les mécanismes fondamentaux de la schizophrénie

Abstract Fundamental research in psychiatric neurosciences assumes that psychiatric disorders are associated with neurobiological factors. Identification of these factors would provide therapeutic targets as well as a better understanding of the relationship between- brain and behaviour in pathological processes. We conducted experiments in an animal model of schizophrenia. Several behavioural tasks were used to evaluate spatial and working memory in these animals. The model is based on glutathione deficit during cerebral development. Indeed, a 50% decrease of glutathione has been reported in prefrontal cortex of patients with schizophrenia. Glutathione is a major antioxidant in the brain and its deficit could lead to abnormal brain connectivity. The glutathione deficit was induced in rats by perinatal (PS-P16) subcutaneous injections with Lbuthionine-(S,R)-sullfoximine (BSO), an inhibitor of glutathione synthesis. This treatment leads to a transitory 50% glutathione levels during brain development. In parallel, we conducted behavioural testing in rats with a medial prefrontal cortex lesion. This allowed us to compare early damage induced by BSO treatment with a focal lesion in adults of a brain area known to present anomalies in schizophrenia. Finally, we conducted a series of experiments in senescent rats to evaluate if cognitive deficits could be related to neurobiological changes. Our results show that an early glutathione deficit provokes cognitive deficits in adulthood. These spatial and working memory deficits resemble the cognitive deficits observed in schizophrenia. The comparison with prefrontal rats revealed that the early brain glutathione deficit provoked more severe cognitive deficits than the prefrontal lesion in adult rats. Moreover, in both cases, we observed a dissociation in memory deficits depending on the type of locomotion that was used in behavioural experiments. Indeed, BSO treated rats as well as prefrontal rats showed place learning or working memory deficits in tasks conducted on dry surfaces where they had to walk. In contrast, they showed no deficit when the same cognitive functions were tested in the water maze. This dissociation might be sustained by a difference in requirement of sensory integration between walking and swimming tasks. Résumé La recherche fondamentale en neurosciences psychiatriques repose sur le présupposé selon lequel les symptômes manifestés dans les troubles psychiatriques auraient des concomitants neurobiologiques. Ceux-ci, une fois identifiés, offriraient des cibles pour une démarche thérapeutique ainsi que des modèles permettant de mieux comprendre les soubassements biologiques du comportement et des activités mentales. Nos expériences s'articulent autour de la question de la modélisation de la schizophrénie chez l'animal. Nous avons recherché chez ces animaux des troubles cognitifs et sensoriels associés à la schizophrénie. En effet, chez l'homme comme chez l'animal, la mémoire spatiale et la mémoire de travail dépendent fortement de la capacité d'intégration et d'organisation des informations sensorielles. Les premières expériences ont été menées suite à une perturbation périnatale du développement cérébral. Celle-ci visait à reproduire une diminution du taux de glutathion dans le cerveau, des recherches précédentes ayant observé une diminution de 50% du taux de glutathion dans le cortex préfrontal de patients schizophrènes. Le glutathion étant un antioxydant majeur dans le cerveau, son déficit pourrait conduire à des perturbations de la circuiterie cérébrale. Nous avons reproduit ce déficit chez le rat, par injection de Lbuthionine-(S,R)-sullfoximine (BSO), un inhibiteur de la synthèse du glutathion... Ce traitement a été administré pendant la période périnatale (du jour postnatal 5 au jour 16) provoquant une diminution de 50% du taux de glutathion. Nous avons ensuite évalué lës répercussions de cette atteinte précoce sur le comportement des rats à l'âge adulte. Ce modèle s'inscrit donc dans l'hypothèse neurodéveloppementale qui associe la schizophrénie à une atteinte du développement cérébral normal. Nous avons ensuite conduit des expériences similaires chez des rats ayant subi une lésion du cortex préfrontal pour comparer les répercussions du traitement périnatal avec une lésion, à l'âge adulte, d'une aire cérébrale connue pour présenter des anomalies chez les patients. Finalement, nous avons évalué si les processus sensoriels et cognitifs précédemment étudiés pouvaient également être affectés lors du vieillissement normal en recherchant des corrélats biologiques des déficits de mémoire liés à l'âge avancé. Nos résultats montrent que ce déficit précoce en glutathion peut avoir des répercussions surale comportement à l'âge adulte. On a relevé une similarité avec les déficits cognitifs associés.à la schizophrénie, incluant des déficits de mémoire de travail ainsi que des déficits de mémoire spatiale. Ces déficits étaient fortement liés au type de locomotion utilisée et n'ont été observés que dans les tâches où les animaux devaient rejoindre un but en marchant mais pas dans lés tests dans lesquels ils devaient localiser une cible en nageant. Les déficits induits par la lésion préfrontale chez l'adulte étaient beaucoup plus légers que ceux découlant de l'atteinte périnatale mais présentaient une dissociation analogue en fonction du type de locomotion. De plus, des tests similaires menés au cours du vieillissement confirment que la mémoire de travail peut être affectée sélectivement par le vieillissement dans une tâche où les animaux doivent marcher, tout en restant intacte dans le bassin de Morris. Les déficits cognitifs liés au vieillissement étaient significativement corrélés à des différences de niveaux des protéines post-synaptiques PSD95 (postsynaptic density 95). L'ensemble des résultats montre que les tests qui sont fréquemment utilisés pour évaluer la mémoire chez l'animal pourraient faire appel à des processus différents. Cette différence pourrait notamment tenir au niveau d'intégration sensorielle requis pour résoudre la tâche, qui est particulièrement sollicitée au cours d'une locomotion intermittente.

Inherently self-calibrating non-Cartesian parallel imaging.

The use of self-calibrating techniques in parallel magnetic resonance imaging eliminates the need for coil sensitivity calibration scans and avoids potential mismatches between calibration scans and subsequent accelerated acquisitions (e.g., as a result of patient motion). Most examples of self-calibrating Cartesian parallel imaging techniques have required the use of modified k-space trajectories that are densely sampled at the center and more sparsely sampled in the periphery. However, spiral and radial trajectories offer inherent self-calibrating characteristics because of their densely sampled center. At no additional cost in acquisition time and with no modification in scanning protocols, in vivo coil sensitivity maps may be extracted from the densely sampled central region of k-space. This work demonstrates the feasibility of self-calibrated spiral and radial parallel imaging using a previously described iterative non-Cartesian sensitivity encoding algorithm.

Strategies for sequential prediction of stationary time series

We present simple procedures for the prediction of a real valued sequence. The algorithms are based on a combinationof several simple predictors. We show that if the sequence is a realization of a bounded stationary and ergodic random process then the average of squared errors converges, almost surely, to that of the optimum, given by the Bayes predictor. We offer an analog result for the prediction of stationary gaussian processes.

On a series of Goldbach and Euler

Theorem 1 of Euler s paper of 1737 'Variae Observationes Circa Series Infinitas', states the astonishing result that the series of all unit fractions whose denominators are perfect powers of integers minus unity has sum one. Euler attributes the Theorem to Goldbach. The proof is one of those examples of misuse of divergent series to obtain correct results so frequent during the seventeenth and eighteenth centuries. We examine this proof closelyand, with the help of some insight provided by a modern (and completely dierent) proof of the Goldbach-Euler Theorem, we present a rational reconstruction in terms which could be considered rigorous by modern Weierstrassian standards. At the same time, with a few ideas borrowed from nonstandard analysis we see how the same reconstruction can be also be considered rigorous by modern Robinsonian standards. This last approach, though, is completely in tune with Goldbach and Euler s proof. We hope to convince the reader then how, a few simple ideas from nonstandard analysis, vindicate Euler's work.

Improved nonparametric confidence intervals in time series regressions

Condence intervals in econometric time series regressions suffer fromnotorious coverage problems. This is especially true when the dependencein the data is noticeable and sample sizes are small to moderate, as isoften the case in empirical studies. This paper suggests using thestudentized block bootstrap and discusses practical issues, such as thechoice of the block size. A particular data-dependent method is proposedto automate the method. As a side note, it is pointed out that symmetricconfidence intervals are preferred over equal-tailed ones, since theyexhibit improved coverage accuracy. The improvements in small sampleperformance are supported by a simulation study.

Human large granular lymphocytes contain an esterase activity usually considered as specific for the myeloid series.

A cytochemical marker such as alpha-naphthyl acetate esterase (ANAE) has been found useful for the morphological identification of the subset of T lymphocytes having receptors for Fcμ (TM cells). ANAE reaction on TM cells gives a typical pattern of one to four positive spots, whereas this pattern is not found on T cells with receptors for Fcγ (TG cells). ANAE is abundant in monocytes but not detectable in granulocytes. Herein another type of esterase activity, naphthol-AS-D chloroacetate esterase (NCAE), is described; it is known to be abundant in granulocytes and was found to give a specific pattern of reactivity with the subpopulation of large granular lymphocytes (LGL). This pattern of fine granular staining was observed not only on LGL present in the TG cell subpopulation but also in LGL present in the non-T, non-B cells. Fractions of peripheral blood mononuclear cells which were ènriched up to 80% in LGL by Percoll discontinuous density gradient gave a similar percentage of specific NCAE pattern. In addition, among the different fractions from Percoll gradient, there was a good correlation (r = 0.94) between the number of NCAE-positive cells and the natural killer activity against the natural killer susceptible K562 target cells. It will be important to determine whether or not this enzymatic activity plays a role in the cytotoxic activities of LGL.

Parallel scheduling of multiclass M/M/m queues: Approximate and heavy-traffic optimization of achievable performance

We address the problem of scheduling a multiclass $M/M/m$ queue with Bernoulli feedback on $m$ parallel servers to minimize time-average linear holding costs. We analyze the performance of a heuristic priority-index rule, which extends Klimov's optimal solution to the single-server case: servers select preemptively customers with larger Klimov indices. We present closed-form suboptimality bounds (approximate optimality) for Klimov's rule, which imply that its suboptimality gap is uniformly bounded above with respect to (i) external arrival rates, as long as they stay within system capacity;and (ii) the number of servers. It follows that its relativesuboptimality gap vanishes in a heavy-traffic limit, as external arrival rates approach system capacity (heavy-traffic optimality). We obtain simpler expressions for the special no-feedback case, where the heuristic reduces to the classical $c \mu$ rule. Our analysis is based on comparing the expected cost of Klimov's ruleto the value of a strong linear programming (LP) relaxation of the system's region of achievable performance of mean queue lengths. In order to obtain this relaxation, we derive and exploit a new set ofwork decomposition laws for the parallel-server system. We further report on the results of a computational study on the quality of the $c \mu$ rule for parallel scheduling.