Abstract
In this article, the problem of deriving a physical model of a mechanical structure from an arbitrary state-space realization is addressed. As an alternative to finite element formulations, the physical parameters of a model may be directly obtained from identified parametric models. However, these methods are limited by the number of available sensors and often lead to poor predictive models. Additionally, the most efficient identification algorithms retrieve models where the physical parameters are hidden. This last difficulty is known in the literature as the inverse vibration problem. In this work, an approach to the inverse vibration problem is proposed. It is based on a similarity transformation and the requirement that every degree of freedom should contain a sensor and an actuator (full instrumented system) is relaxed to a sensor or an actuator per degree of freedom, with at least one co-located pair (partially instrumented system). The physical parameters are extracted from a state-space realization of the former system. It is shown that this system has a symmetric transfer function and this symmetry is exploited to derive a state-space realization from an identified model of the partially instrumented system. A subspace continuous-time system identification algorithm previously proposed by the authors in [1] is used to estimate this model from the IO data.

Abstract
The emergence of ubiquitous sources of streaming data has given rise to the popularity of algorithms for online machine learning. In that context, Hoeffding trees represent the state-of-the-art algorithms for online classification. Their popularity stems in large part from their ability to process large quantities of data with a speed that goes beyond the processing power of any other streaming or batch learning algorithm. As a consequence, Hoeffding trees have often been used as base models of many ensemble learning algorithms for online classification. However, despite the existence of many algorithms for online classification, ensemble learning algorithms for online regression do not exist. In particular, the field of online any-time regression analysis seems to have experienced a serious lack of attention. In this paper, we address this issue through a study and an empirical evaluation of a set of online algorithms for regression, which includes the baseline Hoeffding-based regression trees, online option trees, and an online least mean squares filter. We also design, implement and evaluate two novel ensemble learning methods for online regression: online bagging with Hoeffding-based model trees, and an online RandomForest method in which we have used a randomized version of the online model tree learning algorithm as a basic building block. Within the study presented in this paper, we evaluate the proposed algorithms along several dimensions: predictive accuracy and quality of models, time and memory requirements, bias-variance and bias-variance-covariance decomposition of the error, and responsiveness to concept drift.

Abstract
IEEE 1149.1, commonly known as the joint test action group (JTAG), is the standard for the test access port and the boundary-scan architecture. The JTAG is primarily utilized at the time of the integrated circuit (IC) manufacture but also in the field, giving access to internal sub-systems of the IC, or for failure analysis and debugging. Because the JTAG needs to be left intact and operational for use, it inevitably provides a "backdoor" that can be exploited to undermine the security of the chip. Potential attackers can then use the JTAG to dump critical data or reverse engineer IP cores, for example. Since an attacker will use the JTAG differently from a legitimate user, it is possible to detect the difference using machine-learning algorithms. A JTAG protection scheme, SLIC-J, is proposed to monitor user behavior and detect illegitimate accesses to the JTAG. Specifically, JTAG access is characterized using a set of specifically-defined features, and then an on-chip classifier is used to predict whether the user is legitimate or not. To validate the effectiveness of the approach, both legitimate and illegitimate JTAG accesses are simulated using the OpenSPARC T2 benchmark. The results show that the detection accuracy is 99.2%, and the escape rate is 0.8%.

Abstract

Abstract
The emergence of the so-called “European Paradox” shows that increasing Governmental R&D Investment is far from being a ‘panacea’ for stagnant growth. It is worth noting that Governmental R&D Investment does not have a statistically significant impact on employment, indicating the need to assess the trade-offs of policies that could lead to significant increases in government expenditure. Surprisingly, Governmental R&D Employment does not contribute to ‘mass-market’ employment, despite its quite important role in reducing Youth-Unemployment. Despite the negative side-effects of Governmental R&D Employment on both GVA and GDP, University R&D Employment appears to have a quite important role in reducing Unemployment, especially Youth-Unemployment, while it also does not have a downside in terms of economic growth. Technological Capacity enhancement is the most effective instrument for reducing Unemployment and is a policy without any downside regarding sustainable economic development. © Universidad Alberto Hurtado, Facultad de Economía y Negocios.

Abstract
Developments in advanced autonomous production resources have increased the interest in the Single-Machine Scheduling Problem (SMSP). Until now, researchers used SMSP with little to no practical application in industry, but with the introduction of multi-purpose machines, able of executing an entire task, such as 3D Printers, replacing extensive production chains, single-machine problems are becoming a central point of interest in real-world scheduling. In this paper we study how simple, easy to implement, Just-in-Time (JIT) based, constructive heuristics, can be used to optimize customer and enterprise oriented performance measures. Customer oriented performance measures are mainly related to the accomplishment of due dates while enterprise-oriented ones typically consider other time-oriented measures. © 2015 AISTI.