Abstract
Design and hierarchical control of three phase parallel Voltage Source Inverters are developed in this paper. The control scheme is based on synchronous reference frame and consists of primary and secondary control levels. The primary control consists of the droop control and the virtual output impedance loops. This control level is designed to share the active and reactive power correctly between the connected VSIs in order to avoid the undesired circulating current and overload of the connected VSIs. The secondary control is designed to clear the magnitude and the frequency deviations caused by the primary control. The control structure is validated through dynamics simulations. The obtained results demonstrate the effectiveness of the control structure.

Abstract
Stochastic search algorithms are black-box optimizer of an objective function. They have recently gained a lot of attention in operations research, machine learning and policy search of robot motor skills due to their ease of use and their generality. However, when the task or objective function slightly changes, many stochastic search algorithms require complete re-leaming in order to adapt thesolution to the new objective function or the new context. As such, we consider the contextual stochastic search paradigm. Here, we want to find good parameter vectors for multiple related tasks, where each task is described by a continuous context vector. Hence, the objective function might change slightly for each parameter vector evaluation. In this paper, we investigate a contextual stochastic search algorithm known as Contextual Relative Entropy Policy Search (CREPS), an information-theoretic algorithm that can learn from multiple tasks simultaneously. We show the application of CREPS for simulated robotic tasks.

Abstract
We describe different scenarios where a learning mechanism is capable of acquiring experience on a source task, and subsequently exploit such experience on a target task. The core ideas behind this ability to transfer knowledge from one task to another have been studied in the machine learning literature under different titles and perspectives. Here we describe some of them under the names of inductive transfer, transfer learning, multitask learning, meta-searching, meta-generalization, and domain adaptation.

Abstract

Abstract
The ageing process is naturally accompanied by changes in people's cognitive processes. The European population ageing is a challenge for the European social policy and for the mental health professionals. New technologies can play an important role in the neurocognitive stimulation area as they possess characteristics that might reduce the anxiety levels of patients participating in neurocognitive stimulation or assessment programs. In particular, serious games provide a setting that can be explored to improve the easy access to neurocognitive stimulation and assessment, regardless of place and time, at a lower cost then traditional approaches. This paper presents a serious game aiming to analyse neurocognitive deficits and stimulate the players' deficitary neurocognitive processes. This game is built on top of sound neurocognitive psychotherapy for adults, mainly addressing the cognitive processes of attention and memory. The game will simulate real world scenarios, allowing a better generalization process due to ecological validity.

Abstract