Abstract
A fundamental task for artificial intelligence is learning. Deep Neural Networks have proven to cope perfectly with all learning paradigms, i.e. supervised, unsupervised, and reinforcement learning. Nevertheless, traditional deep learning approaches make use of cloud computing facilities and do not scale well to autonomous agents with low computational resources. Even in the cloud, they suffer from computational and memory limitations, and they cannot be used to model adequately large physical worlds for agents which assume networks with billions of neurons. These issues are addressed in the last few years by the emerging topic of sparse training, which trains sparse networks from scratch. This paper discusses sparse training state-of-the-art, its challenges and limitations while introducing a couple of new theoretical research directions which has the potential of alleviating sparse training limitations to push deep learning scalability well beyond its current boundaries. Nevertheless, the theoretical advancements impact in complex multi-agents settings is discussed from a real-world perspective, using the smart grid case study.

Abstract
Nowadays we are facing the emergence of new challenges, especially focused on environmental behavior and climate change and also the effects and impacts of COVID-19 world pandemic - which can restrain the attainment of the desired sustainability. It is, then, mandatory to reply to all of these challenges through the design of specific paths to attain sustainable development in a collective approach, with the involvement and the commitment of the community and performed in an integrated way. Thus, it is proposed a trans-disciplinary research method based on the European Union title recognition - European Green Capital (EGC) -, directly related to the Circular Economy (CE), together with a Data Envelopment Analysis (DEA). This paper highlights sustainable measures and proposes common managerial strategies and policies, that can support the cities/regions' sustainable practices embedding as well as ensure its overall monitoring to measure if the actions implemented through time are adequate and efficient towards attaining CE and sustainable development. Those are based on the analysis of the best practices of the EGC, which can impact on CE, and using DEA. This approach can follow the city/region evolution and be adapted to the EGC evaluation parameters in order to understand the main characteristics that can contribute to improve governance approaches and help to foster CE into all the cities/regions' ecosystem.

Abstract
Twitter has become a major platform to share ideas and promoting discussion on relevant topics. However, with a large number of users to resort to it as their primary source of information and with an increasing number of accounts spreading newsworthy content, a characterization of the political bias associated with the social network ecosystem becomes necessary. In this work, we aim at analyzing accounts spreading or publishing content from five different classes of the political spectrum. We also look further and study accounts who spread content from both right and left sides. Conclusions show that there is a large presence of accounts which disseminate right bias content although it is the more central classes that have a higher influence on the network. In addition, users who spread content from both sides are more actively spreading right content with opposite content associated with criticism towards left political parties or promoting right political decisions.

Abstract
System flexibility has been introduced as one of the most significant concepts in energy systems, and accordingly it has captured attention. It should be noted that various parameters and equipment, directly and indirectly, affect system flexibility, among which, demand response (DR) programs, distributed energy resources (DERs), and storage systems, are some important examples. In this respect, a comprehensive review of DR and integrated demand response (IDR) programs has been conducted in this chapter, and the impact of such programs on enhancing the flexibility of local energy systems has been thoroughly investigated. The local energy systems, studied in this chapter, include three residential, commercial, and industrial energy hubs, located in a 33-bus network, equipped with renewable energy sources (RES), as well as electrical and thermal energy storage systems. It should be noted that to evaluate the flexibility of the system, the operation problem of energy hubs has been investigated through simulating six different case studies, and the impact of DR/IDR programs, energy storage systems, RESs, and operation mode has been evaluated on operating costs, emissions, and flexibility. The results showed that each of the hubs will have a different reaction to the presence/absence of the mentioned items.

Abstract
The task of person re-identification has important applications in security and surveillance systems. It is a challenging problem since there can be a lot of differences between pictures belonging to the same person, such as lighting, camera position, variation in poses and occlusions. The use of Deep Learning has contributed greatly towards more effective and accurate systems. Many works use attention mechanisms to force the models to focus on less distinctive areas, in order to improve performance in situations where important information may be missing. This paper proposes a new, more flexible method for calculating these masks, using a U-Net which receives a picture and outputs a mask representing the most distinctive areas of the picture. Results show that the method achieves an accuracy comparable or superior to those in state-of-the-art methods.

Abstract
Several factors have been identified to contribute to the sense of presence and cybersickness, including the preponderance users have in the virtual environments (VE) and their gender. This work focuses on studying the Role Type and gender in a VE and their impact on the sense of presence and cybersickness when immersive Virtual Reality (VR) setups are used. For this, a set of psychophysical experiments were conducted to evaluate a VR scenario with three Role Types: Viewer, Explorer, and Searcher. Results revealed statistically significant differences in Spatial Presence, Cybersickness, Nausea, Oculomotor Discomfort, and Disorientation for Role Type. In the evaluated scenario, it was observed that a more dominant Role Type on the VE leads to a higher reported spatial presence (sense of physically being present in the VE) and higher cybersickness scores. We conclude that a higher relevance of the Role Type makes the users more sensitive to the stimuli present in the VE regarding the coherency of the interaction/simulation and, consequently, more prone to develop cybersickness symptoms. No differences were found between the genders.