Abstract
https://aimc2023.pubpub.org/pub/latent-spaces-tonal-music Variational Autoencoders (VAEs) have proven to be effective models for producing latent representations of cognitive and semantic value. We assess the degree to which VAEs trained on a prototypical tonal music corpus of 371 Bach's chorales define latent spaces representative of the circle of fifths and the hierarchical relation of each key component pitch as drawn in music cognition. In detail, we compare the latent space of different VAE corpus encodings — Piano roll, MIDI, ABC, Tonnetz, DFT of pitch, and pitch class distributions — in providing a pitch space for key relations that align with cognitive distances. We evaluate the model performance of these encodings using objective metrics to capture accuracy, mean square error (MSE), KL- divergence, and computational cost. The ABC encoding performs the best in reconstructing the original data, while the Pitch DFT seems to capture more information from the latent space. Furthermore, an objective evaluation of 12 major or minor transpositions per piece is adopted to quantify the alignment of 1) intra- and inter-segment distances per key and 2) the key distances to cognitive pitch spaces. Our results show that Pitch DFT VAE latent spaces align best with cognitive spaces and provide a common-tone space where overlapping objects within a key are fuzzy clusters, which impose a well-defined order of structural significance or stability — i.e., a tonal hierarchy. Tonal hierarchies of different keys can be used to measure key distances and the relationships of their in-key components at multiple hierarchies (e.g., notes and chords). The implementation of our VAE and the encodings framework are made available online.

Abstract
Among the seven key requirements to achieve trustworthy AI proposed by the High-Level Expert Group on Artificial Intelligence (AI-HLEG) established by the European Commission, the fifth requirement (“Diversity, non-discrimination and fairness”) declares: “In order to achieve Trustworthy AI, we must enable inclusion and diversity throughout the entire AI system’s life cycle. [...] This requirement is closely linked with the principle of fairness”. In this paper, we try to shed light on how closely these two distinct concepts, diversity and fairness, may be treated by focusing on information access systems and ranking literature.

Abstract
Recent societal changes have brought new challenges to contemporary organisations, e.g., how to properly manage the work-family/family-work dyad and, thus, promote adequate task performance. This paper aimed to study the relationship between conflict (work-family and family-work) and task performance, and whether this relationship was moderated by well-being. Thus, the following hypotheses were formulated: (1) conflict (work-family and family-work) is negatively associated with task performance; (2) conflict (work-family and family-work) is negatively associated with well-being; (3) well-being is positively associated with task performance; and (4) well-being moderates the relationship between conflict (work-family and family-work) and task performance. A total of 596 subjects participated in this study, all employed in Portuguese organisations. The results underlined that only family-work conflict was negatively and significantly associated with task performance. Work-family conflict established a negative and significant relationship with well-being. Well-being was positively and significantly associated with performance and moderated the relationship between conflict (work-family and family-work) and task performance. These results show that organisations should provide employees with situations that promote their well-being, especially in Portugal, where a relationship culture exists (rather than task culture, which is predominant in the USA and Canada, for example) which means that additional and considerable time must be dedicated to personal and family matters for people to fit in and be accepted harmoniously.

Abstract
The growing demand for conventional internal combustion engine vehicles can aggravate the current energy and environmental crisis by presenting a higher dependence on fossil fuels and a higher level of greenhouse gases. The transition to electric mobility is a sustainable solution to mitigate the negative impact on the environment and energy security. In recent years, global sales of electric vehicles (EVs) have grown steadily; however, their worldwide market share is still less than 10%. The present study aims to improve and strengthen the knowledge base on consumer behavior toward EV purchases by investigating the antecedents of EV purchase intention, as well as their impact. This paper is based on a systematic literature review where 63 articles published between 1994 and 2021 were analyzed. The antecedents were classified into three main categories: consumer characteristics, EV characteristics, and EV-related policies. A summary model represents the impact information of each of the main antecedents. The descriptive results of the sample are also discussed. Finally, this study indicates directions for future research and recommendations for developing the most effective strategies and policies that will accelerate the transition to a more sustainable future.

Abstract
In recent years growing volumes of data have made the task of applying various machine learning algorithms a challenge in a great number of cases. This challenge is posed in two main ways: training time and processing load. Normally, problems in these two categories may be attributed to irrelevant, redundant, or noisy features. So as to avoid this type of feature most pre-processing pipelines include a step dedicated so selecting the most relevant features or combining existing ones into a single better representation. These techniques are denominated dimensionality reduction techniques. In this work, we aim to present a short look at the current state of the art in this area. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract
Many real-world datasets have an underlying dynamic graph structure, where entities and their interactions evolve over time. Machine learning models should consider these dynamics in order to harness their full potential in downstream tasks. Previous approaches for graph representation learning have focused on either sampling khop neighborhoods, akin to breadth-first search, or random walks, akin to depth-first search. However, these methods are computationally expensive and unsuitable for real-time, low-latency inference on dynamic graphs. To overcome these limitations, we propose graph-sprints a general purpose feature extraction framework for continuous-time-dynamic-graphs (CTDGs) that has low latency and is competitive with state-of-the-art, higher latency models. To achieve this, a streaming, low latency approximation to the random-walk based features is proposed. In our framework, time-aware node embeddings summarizing multi-hop information are computed using only single-hop operations on the incoming edges. We evaluate our proposed approach on three open-source datasets and two in-house datasets, and compare with three state-of-the-art algorithms (TGN-attn, TGN-ID, Jodie). We demonstrate that our graph-sprints features, combined with a machine learning classifier, achieve competitive performance (outperforming all baselines for the node classification tasks in five datasets). Simultaneously, graphsprints significantly reduce inference latencies, achieving close to an order of magnitude speed-up in our experimental setting.