RESUMO
For the global COVID-19 pandemic it is still not adequately understood how quarantine disobedience and change in mobility restrictions influence the pandemic spreading and waves. Here, we propose a new metapopulation epidemiological model as a network composed of equal clusters to predict the course of the epidemic based on the contiguous spreading between the neighbours, the probability of quarantine misbehaviour, and the probability of mobility, which control contacts outside the cluster. We exemplify the model by comparing simulation results with real data on COVID-19 pandemic in Croatia. Fitting the data over the first and second pandemic waves, when the probability of mobility is set by the stringency index, the probability of quarantine misbehaviour is found by a Bayesian optimization yielding a fascinating agreement between the daily COVID-19 deaths and model output and efficiently predicting the timing of pandemic bursts. A sudden increase in the probability of quarantine misbehaviour alongside the sudden increase in the probability of mobility generate the model third wave in good agreement with daily COVID-19 deaths.
RESUMO
Well-evidenced advances of data-driven complex machine learning approaches emerging within the so-called second wave of artificial intelligence (AI) fostered the exploration of possible AI applications in various domains and aspects of human life, practices, and society [...].
RESUMO
The adaptation of deep learning models within safety-critical systems cannot rely only on good prediction performance but needs to provide interpretable and robust explanations for their decisions. When modeling complex sequences, attention mechanisms are regarded as the established approach to support deep neural networks with intrinsic interpretability. This paper focuses on the emerging trend of specifically designing diagnostic datasets for understanding the inner workings of attention mechanism based deep learning models for multivariate forecasting tasks. We design a novel benchmark of synthetically designed datasets with the transparent underlying generating process of multiple time series interactions with increasing complexity. The benchmark enables empirical evaluation of the performance of attention based deep neural networks in three different aspects: (i) prediction performance score, (ii) interpretability correctness, (iii) sensitivity analysis. Our analysis shows that although most models have satisfying and stable prediction performance results, they often fail to give correct interpretability. The only model with both a satisfying performance score and correct interpretability is IMV-LSTM, capturing both autocorrelations and crosscorrelations between multiple time series. Interestingly, while evaluating IMV-LSTM on simulated data from statistical and mechanistic models, the correctness of interpretability increases with more complex datasets.
RESUMO
Machines usually employ a guess-and-check strategy to analyze data: they take the data, make a guess, check the answer, adjust it with regard to the correct one if necessary, and try again on a new data set. An active learning environment guarantees better performance while training on less, but carefully chosen, data which reduces the costs of both annotating and analyzing large data sets. This issue becomes even more critical for deep learning applications. Human-like active learning integrates a variety of strategies and instructional models chosen by a teacher to contribute to learners' knowledge, while machine active learning strategies lack versatile tools for shifting the focus of instruction away from knowledge transmission to learners' knowledge construction. We approach this gap by considering an active learning environment in an educational setting. We propose a new strategy that measures the information capacity of data using the information function from the four-parameter logistic item response theory (4PL IRT). We compared the proposed strategy with the most common active learning strategies-Least Confidence and Entropy Sampling. The results of computational experiments showed that the Information Capacity strategy shares similar behavior but provides a more flexible framework for building transparent knowledge models in deep learning.
RESUMO
The complexity of the cerebral cortex underlies its function and distinguishes us as humans. Here, we present a principled veridical data science methodology for quantitative histology that shifts focus from image-level investigations towards neuron-level representations of cortical regions, with the neurons in the image as a subject of study, rather than pixel-wise image content. Our methodology relies on the automatic segmentation of neurons across whole histological sections and an extensive set of engineered features, which reflect the neuronal phenotype of individual neurons and the properties of neurons' neighborhoods. The neuron-level representations are used in an interpretable machine learning pipeline for mapping the phenotype to cortical layers. To validate our approach, we created a unique dataset of cortical layers manually annotated by three experts in neuroanatomy and histology. The presented methodology offers high interpretability of the results, providing a deeper understanding of human cortex organization, which may help formulate new scientific hypotheses, as well as to cope with systematic uncertainty in data and model predictions.
Assuntos
Córtex Cerebral , Aprendizado de Máquina , Humanos , Córtex Cerebral/anatomia & histologia , Neurônios , Processamento de Imagem Assistida por ComputadorRESUMO
Physics has a long tradition of laying rigorous quantitative foundations for social phenomena. Here, we up the ante for physics' forays into the territory of social sciences by (i) empirically documenting a tipping point in the relationship between democratic norms and corruption suppression, and then (ii) demonstrating how such a tipping point emerges from a micro-scale mechanistic model of spin dynamics in a complex network. Specifically, the tipping point in the relationship between democratic norms and corruption suppression is such that democratization has little effect on suppressing corruption below a critical threshold, but a large effect above the threshold. The micro-scale model of spin dynamics underpins this phenomenon by reinterpreting spins in terms of unbiased (i.e. altruistic) and biased (i.e. parochial) other-regarding behaviour, as well as the corresponding voting preferences. Under weak democratic norms, dense social connections of parochialists enable coercing enough opportunist voters to vote in favour of perpetuating parochial in-group bias. Society may, however, strengthen democratic norms in a rapid turn of events during which opportunists adopt altruism and vote to subdue bias. The emerging model outcome at the societal scale thus mirrors the data, implying that democracy either perpetuates or suppresses corruption depending on the prevailing democratic norms.
RESUMO
Spike glycoprotein is essential for the reproduction of the SARS-CoV-2 virus, and its inhibition using already approved antiviral drugs may open new avenues for treatment of patients with the COVID-19 disease. Because of that we analyzed the inhibition of SARS-CoV-2 spike glycoprotein with FDA-approved antiviral drugs and their double and triple combinations. We used the VINI in silico model of cancer to perform this virtual drug screening, showing HIV drugs to be the most effective. Besides, the combination of cobicistat-abacavir-rilpivirine HIV drugs demonstrated the highest in silico efficacy of inhibiting SARS-CoV-2 spike glycoprotein. Therefore, a clinical trial of cobicistat-abacavir-rilpivirine on a limited number of COVID-19 patients in moderately severe and severe condition is warranted.
RESUMO
Recent research in machine learning pointed to the core problem of state-of-the-art models which impedes their widespread adoption in different domains. The models' inability to differentiate between noise and subtle, yet significant variation in data leads to their vulnerability to adversarial perturbations that cause wrong predictions with high confidence. The study is aimed at identifying whether the algorithms inspired by biological evolution may achieve better results in cases where brittle robustness properties are highly sensitive to the slight noise. To answer this question, we introduce the new robust gradient descent inspired by the stability and adaptability of biological systems to unknown and changing environments. The proposed optimization technique involves an open-ended adaptation process with regard to two hyperparameters inherited from the generalized Verhulst population growth equation. The hyperparameters increase robustness to adversarial noise by penalizing the degree to which hardly visible changes in gradients impact prediction. The empirical evidence on synthetic and experimental datasets confirmed the viability of the bio-inspired gradient descent and suggested promising directions for future research. The code used for computational experiments is provided in a repository at https://github.com/yukinoi/bio_gradient_descent.
Assuntos
Aprendizado de Máquina , Modelos TeóricosRESUMO
Estimating the critical points at which complex systems abruptly flip from one state to another is one of the remaining challenges in network science. Due to lack of knowledge about the underlying stochastic processes controlling critical transitions, it is widely considered difficult to determine the location of critical points for real-world networks, and it is even more difficult to predict the time at which these potentially catastrophic failures occur. We analyse a class of decaying dynamic networks experiencing persistent failures in which the magnitude of the overall failure is quantified by the probability that a potentially permanent internal failure will occur. When the fraction of active neighbours is reduced to a critical threshold, cascading failures can trigger a total network failure. For this class of network we find that the time to network failure, which is equivalent to network lifetime, is inversely dependent upon the magnitude of the failure and logarithmically dependent on the threshold. We analyse how permanent failures affect network robustness using network lifetime as a measure. These findings provide new methodological insight into system dynamics and, in particular, of the dynamic processes of networks. We illustrate the network model by selected examples from biology, and social science.