Abstract
Real time series sometimes exhibit various types of "irregularities": missing observations, observations collected not regularly over time for practical reasons, observation times driven by the series itself, or outlying observations. However, the vast majority of methods of time series analysis are designed for regular time series only. A particular case of irregularly spaced time series is that in which the sampling procedure over time depends also on the observed values. In such situations, there is stochastic dependence between the process being modelled and the times of the observations. In this work, we propose a model in which the sampling design depends on all past history of the observed processes. Taking into account the natural temporal order underlying available data represented by a time series, then a modelling approach based on evolutionary processes seems a natural choice. We consider maximum likelihood estimation of the model parameters. Numerical studies with simulated and real data sets are performed to illustrate the benefits of this model-based approach.

Abstract

Abstract
Throughout this paper it is presented a novel elastic joint configuration, being compared with other similar joints found in recent literature. It is presented its modeling, being its estimation process developed offline, based on a proposed experimental setup. This setup enables to monitor and collect data from an absolute encoder and a load cell. Some data obtained from these sensors is then graphically represented, like angle and torque, obtaining some parameters. Finally, through an optimization process, where the error of the angle is minimized, the remaining parameters of the joint are estimated, thus obtaining a realistic model of the system. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

Abstract
Due to convenience and usability, many deep learning (DL) jobs resort to the available shared parallel file system (PFS) for storing and accessing training data when running in HPC environments. Under such a scenario, however, where multiple I/O-intensive applications operate concurrently, the PFS can quickly get saturated with simultaneous storage requests and become a critical performance bottleneck, leading to throughput variability and performance loss. We present MONARCH, a framework-agnostic middleware for hierarchical storage management. This solution leverages the existing storage tiers present at modern supercomputers (e.g., compute node's local storage, PFS) to improve DL training performance and alleviate the current I/O pressure of the shared PFS. We validate the applicability of our approach by developing and integrating an early prototype with the TensorFlow DL framework. Results show that MONARCH can reduce I/O operations submitted to the shared PFS by up to 45%, decreasing training time by 24% and 12%, for I/O-intensive models, namely LeNet and AlexNet.

Abstract
Transformer-based language models have taken many fields in NLP by storm. BERT and its derivatives dominate most of the existing evaluation benchmarks, including those for Word Sense Disambiguation (WSD), thanks to their ability in capturing context-sensitive semantic nuances. However, there is still little knowledge about their capabilities and potential limitations in encoding and recovering word senses. In this article, we provide an in-depth quantitative and qualitative analysis of the celebrated BERT model with respect to lexical ambiguity. One of the main conclusions of our analysis is that BERT can accurately capture high-level sense distinctions, even when a limited number of examples is available for each word sense. Our analysis also reveals that in some cases language models come close to solving coarse grained noun disambiguation under ideal conditions in terms of availability of training data and computing resources. However, this scenario rarely occurs in real-world settings and, hence, many practical challenges remain even in the coarse-grained setting. We also perform an in-depth comparison of the two main language model-based WSD strategies, namely, fine-tuning and feature extraction, finding that the latter approach is more robust with respect to sense bias and it can better exploit limited available training data. In fact, the simple feature extraction strategy of averaging contextualized embeddings proves robust even using only three training sentences per word sense, with minimal improvements obtained by increasing the size of this training data.

Abstract
IntroductionPeople with schizophrenia experience higher levels of stigma compared with other diseases. The analysis of social media content is a tool of great importance to understand the public opinion toward a particular topic.ObjectivesThe aim of this study is to analyse the content of social media on schizophrenia and the most prevalent sentiments towards this disorder.MethodsTweets were retrieved using Twitter’s Application Programming Interface and the keyword “schizophrenia”. Parameters were set to allow the retrieval of recent and popular tweets on the topic and no restrictions were made in terms of geolocation. Analysis of 8 basic emotions (anger, anticipation, disgust, fear, joy, sadness, surprise, and trust) was conducted automatically using a lexicon-based approach and the NRC Word-Emotion Association Lexicon.ResultsTweets on schizophrenia were heterogeneous. The most prevalent sentiments on the topic were mainly negative, namely anger, fear, sadness and disgust. Qualitative analyses of the most retweeted posts added insight into the nature of the public dialogue on schizophrenia.ConclusionsAnalyses of social media content can add value to the research on stigma toward psychiatric disorders. This tool is of growing importance in many fields and further research in mental health can help the development of public health strategies in order to decrease the stigma towards psychiatric disorders.