Corpos de resistência Brugnara, Gisela de Andrade
The ESPecialist,
03/2024, Letnik:
45, Številka:
2
Journal Article
Recenzirano
Na Alta Amazônia ao sul do Solimões, entre vales, rios e cidades, observa-se uma prática contemporânea no uso da indumentária étnica por alguns grupos indígenas, em sua vida de relações urbanas e ...multilocais extra-aldeia, como atitude expressiva e comunicadora de múltiplos sentidos. Tal atitude compõe atualmente um contexto de deslocamentos em rede floresta-cidade-floresta, que portam dinâmicas de resistência cultural e promovem a expansão de territorialidades nativas no espaço da cidade. Neste espaço hostil, a floresta então se materializa através de corpos adornados por cores e desenhos compostos em sintaxe ininteligível ao português, desinvisibilizando o que os processos de dominação tentam apagar.
Kayne (2022) has proposed that the asymmetry of syntax be built into the fundamental operation of Merge itself. This squib reviews some of his proposals and supporting evidence. Departing from Kayne, ...this squib hypothesized that the asymmetric patterns mainly lie in the functional domain of syntax, and the lexical domain may remain symmetric either within a language or cross-linguistically. The Functional Asymmetry Hypothesis (FAH) is supported by the global symmetry of the VO/OV word order, the commutative conjunction structures in the lexical domain, and the free ordering of event-internal adverbs. If the observation is on the right track, it suggests that the asymmetry of syntax, while empirically robust, cannot be entirely reduced to the operation Merge.
The aim of our study was to investigate whether the developmental trajectories of syntax comprehension of participants with or without intellectual disability are comparable. We obtained results from ...a syntax comprehension test of 615 typically developing participants (mean chronological age = 4.61 years, SD = 0.85) and 615 participants with intellectual disability (mean chronological age = 12.22 years, SD = 3.14) matched on a nonverbal cognitive measure. We examined these results by applying several statistical approaches to the overall scores for the syntax test and then to the scores for each of its 92 items. Results showed negligible between‐group differences in developmental trajectories, at both levels of analysis (overall and item scores). This lack of effect suggests that the relationship between cognitive development and syntax comprehension is comparable for the two groups of participants. Theoretical and practical implications of these findings are discussed.
Predicting protein function and structure from sequence is one important challenge for computational biology. For 26 years, most state-of-the-art approaches combined machine learning and evolutionary ...information. However, for some applications retrieving related proteins is becoming too time-consuming. Additionally, evolutionary information is less powerful for small families, e.g. for proteins from the Dark Proteome. Both these problems are addressed by the new methodology introduced here.
We introduced a novel way to represent protein sequences as continuous vectors (embeddings) by using the language model ELMo taken from natural language processing. By modeling protein sequences, ELMo effectively captured the biophysical properties of the language of life from unlabeled big data (UniRef50). We refer to these new embeddings as SeqVec (Sequence-to-Vector) and demonstrate their effectiveness by training simple neural networks for two different tasks. At the per-residue level, secondary structure (Q3 = 79% ± 1, Q8 = 68% ± 1) and regions with intrinsic disorder (MCC = 0.59 ± 0.03) were predicted significantly better than through one-hot encoding or through Word2vec-like approaches. At the per-protein level, subcellular localization was predicted in ten classes (Q10 = 68% ± 1) and membrane-bound were distinguished from water-soluble proteins (Q2 = 87% ± 1). Although SeqVec embeddings generated the best predictions from single sequences, no solution improved over the best existing method using evolutionary information. Nevertheless, our approach improved over some popular methods using evolutionary information and for some proteins even did beat the best. Thus, they prove to condense the underlying principles of protein sequences. Overall, the important novelty is speed: where the lightning-fast HHblits needed on average about two minutes to generate the evolutionary information for a target protein, SeqVec created embeddings on average in 0.03 s. As this speed-up is independent of the size of growing sequence databases, SeqVec provides a highly scalable approach for the analysis of big data in proteomics, i.e. microbiome or metaproteome analysis.
Transfer-learning succeeded to extract information from unlabeled sequence databases relevant for various protein prediction tasks. SeqVec modeled the language of life, namely the principles underlying protein sequences better than any features suggested by textbooks and prediction methods. The exception is evolutionary information, however, that information is not available on the level of a single sequence.
•We present the first longitudinal study of syntactic priming.•We observed different development trajectories for abstract priming and the lexical boost.•The results support dual mechanism accounts ...of priming where abstract priming reflects implicit learning.
Theories of language acquisition vary significantly in their assumptions regarding the content of children’s early syntactic representations and how they subsequently develop towards the adult state. An important methodological tool in tapping syntactic knowledge is priming. In the current paper, we report the first longitudinal investigation of syntactic priming in children, to test the competing predictions of three different theoretical accounts. A sample of 106 children completed a syntactic priming task testing the English active/passive alternation every six months from 36 months to 54 months of age. We tracked both the emergence and development of the abstract priming effect and lexical boost effect. The lexical boost effect emerged late and increased in magnitude over development, whilst the abstract priming effect emerged early and, in a subsample of participants who produced at least one passive at 36 months, decreased in magnitude over time. In addition, there was substantial variation in the emergence of abstract priming amongst our sample, which was significantly predicted by language proficiency measured six months prior. We conclude that children’s representation of the passive is abstracted early, with lexically dependent priming coming online only later in development. The results are best explained by an implicit learning account of acquisition (Chang, F., Dell, G., S., & Bock, K. 2006. Becoming Syntactic. Psychological Review, 113, 234–272), which induces dynamic syntactic representations from the input that continue to change across developmental time.
This paper proposes a decentralized algorithm for solving a consensus optimization problem defined in a static networked multi-agent system, where the local objective functions have the ...smooth+nonsmooth composite form. Examples of such problems include decentralized constrained quadratic programming and compressed sensing problems, as well as many regularization problems arising in inverse problems, signal processing, and machine learning, which have decentralized applications. This paper addresses the need for efficient decentralized algorithms that take advantages of proximal operations for the nonsmooth terms. We propose a proximal gradient exact first-order algorithm (PG-EXTRA) that utilizes the composite structure and has the best known convergence rate. It is a nontrivial extension to the recent algorithm EXTRA. At each iteration, each agent locally computes a gradient of the smooth part of its objective and a proximal map of the nonsmooth part, as well as exchanges information with its neighbors. The algorithm is "exact" in the sense that an exact consensus minimizer can be obtained with a fixed step size, whereas most previous methods must use diminishing step sizes. When the smooth part has Lipschitz gradients, PG-EXTRA has an ergodic convergence rate of O( 1 /k) in terms of the first-order optimality residual. When the smooth part vanishes, PG-EXTRA reduces to P-EXTRA, an algorithm without the gradients (so no "G" in the name), which has a slightly improved convergence rate at o( 1 /k) in a standard (non-ergodic) sense. Numerical experiments demonstrate effectiveness of PG-EXTRA and validate our convergence results.
•Daily accessed, visible street greenery is quantitatively measured at city scale.•An exploratory tool to map priority streets for potential urban greening efforts.•Google Street View (GSV) images ...and machine learning algorithms are used.•It might be biased if we use urban green cover as the only dominant criterion.
The public benefits of visible street greenery have been well recognised in a growing literature. Nevertheless, this issue was rare to be included into urban greenery and planning practices. As a response to this situation, we proposed an actionable approach for quantifying the daily exposure of urban residents to eye-level street greenery by integrating high resolution measurements on both greenery and accessibility. Google Street View (GSV) images in Singapore were collected and extracted through machine learning algorithms to achieve an accurate measurement on visible greenery. Street networks collected from Open Street Map (OSM) were analysed through spatial design network analysis (sDNA) to quantify the accessibility value of each street. The integration of street greenery and accessibility helps to measure greenery from a human-centred perspective, and it provides a decision-support tool for urban planners to highlight areas with prioritisation for planning interventions. Moreover, the performance between GSV-based street greenery and the urban green cover mapped by remote sensing was compared to justify the contribution of this new measurement. It suggested there was a mismatch between these two measurements, i.e., existing top-down viewpoint through satellites might not be equivalent to the benefits enjoyed by city residents. In short, this analytical approach contributes to a growing trend in integrating large, freely-available datasets with machine learning to inform planners, and it makes a step forward for urban planning practices through focusing on the human-scale measurement of accessed street greenery.
Does second-language (L2) syntactic influence on first-language (L1) reflect long-term changes to L1 syntax or occur only as a result of retrieval difficulties during time-constrained tasks? To ...evaluate L2 influence on L1 representation of animacy constraints (an element at the syntax–semantics interface) and word order (narrow syntax), we asked Korean–English bilingual speakers to judge sentences for grammaticality under both speeded and unspeeded conditions (Study 1) and to choose the more acceptable sentence of pairs that contained one grammatical and one ungrammatical sentence (Study 2). We found evidence for L2 influence on L1 animacy constraints in all cases and potential L2 influence on L1 word order in Study 1. These results indicate that L2 influence on L1 syntax can be observed even in conditions that reduce retrieval difficulty, implicating changes to underlying L1 representations. They also support the notion of greater susceptibility to change at the syntax–semantics interface.
PDF
