Arbitrary Labels Research Articles

Soliton generation and conservation laws for vector light pulses propagating in weakly birefringent waveguides

The principal algebra Lp of coupled nonlinear Schrödinger equations describing the propagation of polarized optical pulses and involving four-wave mixing terms is obtained in terms of the three arbitrary labelling parameters of the system of equations. This algebra is found to be five-dimensional, indicating the richness in symmetries of the system. The most general symmetry group transformation by generators of Lp is found and it is shown that this transformation preserves the boundedness of solutions, and an example of transformation of a soliton solution into a completely new soliton of a different type is presented. Moreover, it is explained how an infinite sequence of bounded solutions can thus be generated, yielding new solitons. Several other important properties of solutions and symmetry group transformations of the system are also demonstrated. As the system of Schrödinger equations under study turns out to be of Lagrange type, conservation laws associated with all variational symmetries of Lp are constructed and interpreted. Some symmetry reductions of the system are also derived.

Saving China and admiring Japan: Cultural traitor Qian Daosun

Abstract Qian Daosun (1887–1966) was imprisoned for collaborating with the Provisional Government in North China under Japanese occupation, and to this day he is labelled as hanjian (traitor). Yet, Qian was first and foremost a cultural literatus, librarian, and an exceptional translator with an in-depth understanding of Japanese culture and languages. This article examines the crucial role that Japan and the Japanese language played for Chinese cultural literati in their quest to save China. It also brings to the forefront the dilemmas and agonizing choices Qian faced in his attempt to promote Sino-Japanese cultural exchange in the midst of war, in particular as a librarian. Wartime libraries are highly contested sites of selection, destruction, censorship, preservation, confiscation, and knowledge production. An added layer of complexity was Japan’s cultural policy in China that promoted Japanese-language collections and governed libraries such as the Beijing1 Modern Science Library where Qian worked. What exacerbated Qian’s dilemmas was his upbringing, which led him to form close personal connections with like-minded Japanese literati. Lastly, this article revisits the hanjian label by comparing Qian’s fate to that of other librarians and returned students of Japan, such as May Fourth writer Lu Xun and patriotic bibliophile Zheng Zhenduo. By deliberately examining May Fourth writers alongside hanjian and Japanese intermediaries, the intention is to dismount arbitrary labels and divisions that have set them apart and against each other in the resistance versus collaboration dichotomy.

Understanding quantum machine learning also requires rethinking generalization

Quantum machine learning models have shown successful generalization performance even when trained with few data. In this work, through systematic randomization experiments, we show that traditional approaches to understanding generalization fail to explain the behavior of such quantum models. Our experiments reveal that state-of-the-art quantum neural networks accurately fit random states and random labeling of training data. This ability to memorize random data defies current notions of small generalization error, problematizing approaches that build on complexity measures such as the VC dimension, the Rademacher complexity, and all their uniform relatives. We complement our empirical results with a theoretical construction showing that quantum neural networks can fit arbitrary labels to quantum states, hinting at their memorization ability. Our results do not preclude the possibility of good generalization with few training data but rather rule out any possible guarantees based only on the properties of the model family. These findings expose a fundamental challenge in the conventional understanding of generalization in quantum machine learning and highlight the need for a paradigm shift in the study of quantum models for machine learning tasks.

A Sociolinguistic Study of Personal Names in Balochi Culture

This paper focuses on the personal names of Balochi culture and regards naming practices as an important aspect of the Baloch culture. Personal names are studied in this paper within the perspective of linguistics anthropology. In this paper, the focus of the study is limited to the names of people living in district Kech of Balochistan province. It is hypothesized that Personal names of Balochi culture are not arbitrary labels, but they have sociocultural tags, functions, and meanings. The date to complete requirement of the objectives have been collected through the voter list and some personal names were obtained from the primary and secondary schools of different cities of district Kech. The qualitative method has been used for analyzing the concerned data. They borrow personal names from different cultures due to religious, political, and ideological reasons however, they have always clothed them in the garments of their national ideology. On this purpose the paper discusses the typology of Balochi names including the names taken from the names of days and months, plants and tree, and circumstantial names. It also discusses about compounding naming practices of Balochi culture which are derived from the combination of two single units of words in Balochi language. The Balochi personal names through classifications and categorization show that the choices of Balochi personal names are based on social and cultural considerations. It was observed in the typology of personal names in Balochi culture which shows that how social belief and cultural practices influence the choices or selections of personal names.

The Socio-Linguistic Implications of Personal Names in Mokpe

Purpose: Most of our Cameroonian national languages are demographically and socio-culturally pressured by the English language. The national languages which were previously used in elementary education and religious activities currently do not have any place in any official domain and thus are limited to home and other few immediate domains. Due to this unequal coexistence of the languages, many national languages are considered endangered, calling for efforts from stakeholders to prevent the death of such a precious cultural heritage. In this line of understanding, the objective of this study was to investigate the semantic aspects, naming criteria, and typology of Mopke personal names. It is a contribution to such initiatives like the Bakweri Language and Literacy committee which aims at documenting the Bakweri language in every possible area and means.
 Methodology: To achieve our objective, data was collected through participant observation, in-depth interview and self-intuition to investigate personal names of the Bakweri also called the Mokpe people of the South West Region of Cameroon, considering naming as an important aspect of the society.
 Findings: As regards the theoretical framework, we looked at Mopke names within the purview of linguistic anthropology considering names as not being arbitrary labels but socio-cultural tags that have socio-cultural functions and meanings. In order to analyse and discuss the patterns observed in the data, the Thematic Analysis technique used to present the typology of the names including family names, circumstantial names, theophorous names and flora and fauna names. The paper further examined the changing nature of Mokpe naming system and practice as dictated by cultural contact mainly with Christian/ western culture.
 Recommendations: The study wraps up with a recommendation for further investigation on issues surrounding naming practices and strategic measures to prevent this important African cultural resource.

The fate of labeled and nonlabeled visual features in working memory.

Visual objects often contain several features. Previous studies showed that verbally labeling a visual feature boosts its retention in a continuous format in visual working memory. Yet, the fate of nonlabeled visual features remained unexplored. One hypothesis is that labeling induces tradeoffs in the allocation of working memory capacity across labeled and nonlabeled features. To test this, we asked participants to memorize multi-feature objects (varying in color, orientation, and spatial frequency), while labeling either its (a) color, (b) orientation, or (c) spatial frequency. To inhibit labeling, they repeated "bababa" aloud in a control condition. At the test, labeled and nonlabeled features were reproduced using a continuous scale. Across four experiments, labeling increased continuous memory for the labeled feature, even when labels were arbitrary. Labeling aftereffects on nonlabeled features were mixed: only sometimes guessing increased. These findings are inconsistent with the hypothesis that labeling induces a capacity-allocation tradeoff. Rather, costs to nonlabeled features accrued when the labeling task was attentionally demanding (e.g., using less familiar or arbitrary labels). We conclude that labeling activates conceptual knowledge, thereby protecting and boosting continuous memory of the labeled feature; yet the attentional demands imposed by labeling itself can lead to the forgetting of nonlabeled features. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

The interpolation phase transition in neural networks: Memorization and generalization under lazy training

Modern neural networks are often operated in a strongly overparametrized regime: they comprise so many parameters that they can interpolate the training set, even if actual labels are replaced by purely random ones. Despite this, they achieve good prediction error on unseen data: interpolating the training set does not lead to a large generalization error. Further, overparametrization appears to be beneficial in that it simplifies the optimization landscape. Here, we study these phenomena in the context of two-layers neural networks in the neural tangent (NT) regime. We consider a simple data model, with isotropic covariates vectors in d dimensions, and N hidden neurons. We assume that both the sample size n and the dimension d are large, and they are polynomially related. Our first main result is a characterization of the eigenstructure of the empirical NT kernel in the overparametrized regime Nd≫n. This characterization implies as a corollary that the minimum eigenvalue of the empirical NT kernel is bounded away from zero as soon as Nd≫n and, therefore, the network can exactly interpolate arbitrary labels in the same regime. Our second main result is a characterization of the generalization error of NT ridge regression including, as a special case, min-ℓ2 norm interpolation. We prove that, as soon as Nd≫n, the test error is well approximated by the one of kernel ridge regression with respect to the infinite-width kernel. The latter is in turn well approximated by the error of polynomial ridge regression, whereby the regularization parameter is increased by a “self-induced” term related to the high-degree components of the activation function. The polynomial degree depends on the sample size and the dimension (in particular on logn/logd).

A Theoretical Approach to Ordinal Classification: Feature Space-Based Definition and Classifier-Independent Detection of Ordinal Class Structures

Ordinal classification (OC) is a sub-discipline of multi-class classification (i.e., including at least three classes), in which the classes constitute an ordinal structure. Applications of ordinal classification can be found, for instance, in the medical field, e.g., with the class labels order, early stage-intermediate stage-final stage, corresponding to the task of classifying different stages of a certain disease. While the field of OC was continuously enhanced, e.g., by designing and adapting appropriate classification models as well as performance metrics, there is still a lack of a common mathematical definition for OC tasks. More precisely, in general, a classification task is defined as an OC task, solely based on the corresponding class label names. However, an ordinal class structure that is identified based on the class labels is not necessarily reflected in the corresponding feature space. In contrast, naturally any kind of multi-class classification task can consist of a set of arbitrary class labels that form an ordinal structure which can be observed in the current feature space. Based on this simple observation, in this work, we present our generalised approach towards an intuitive working definition for OC tasks, which is based on the corresponding feature space and allows a classifier-independent detection of ordinal class structures. To this end, we introduce and discuss novel, OC-specific theoretical concepts. Moreover, we validate our proposed working definition in combination with a set of traditionally ordinal and traditionally non-ordinal data sets, and provide the results of the corresponding detection algorithm. Additionally, we motivate our theoretical concepts, based on an illustrative evaluation of one of the oldest and most popular machine learning data sets, i.e., on the traditionally non-ordinal Fisher’s Iris data set.

Distance-based clustering challenges for unbiased benchmarking studies

Benchmark datasets with predefined cluster structures and high-dimensional biomedical datasets outline the challenges of cluster analysis: clustering algorithms are limited in their clustering ability in the presence of clusters defining distance-based structures resulting in a biased clustering solution. Data sets might not have cluster structures. Clustering yields arbitrary labels and often depends on the trial, leading to varying results. Moreover, recent research indicated that all partition comparison measures can yield the same results for different clustering solutions. Consequently, algorithm selection and parameter optimization by unsupervised quality measures (QM) are always biased and misleading. Only if the predefined structures happen to meet the particular clustering criterion and QM, can the clusters be recovered. Results are presented based on 41 open-source algorithms which are particularly useful in biomedical scenarios. Furthermore, comparative analysis with mirrored density plots provides a significantly more detailed benchmark than that with the typically used box plots or violin plots.

Minimal but not meaningless: Seemingly arbitrary category labels can imply more than group membership.

Minimal group paradigms tend to involve contrived group distinctions, such as dot estimation tendencies and aesthetic preferences. Researchers assume that these novel category distinctions lack informational value. Our research tests this notion. Specifically, we used the classic overestimator versus underestimator and Klee versus Kandinsky minimal group paradigms to assess how category labels influence minimal group responses. In Study 1, we show that participants represented ingroup faces more favorably than outgroup faces, but also represented overestimator and underestimator category labels differently. In fact, the category label effect was larger than the intergroup effect, even though participants were told that estimation tendencies were unrelated to other cognitive tendencies or personality traits. In Study 2, we demonstrate that Klee and Kandinsky were also represented differently, but in this case, the intergroup effect was stronger than the category label effect. In Studies 3 and 4, we examined effects of category labels on how participants allocate resources to, evaluate, and ascribe traits to ingroup and outgroup members. We found both category label and intergroup effects when participants were assigned to overestimator and underestimator groups. However, we found only the intergroup effect when participants were assigned to Klee and Kandinsky groups. Together, this work advances but does not upend understanding of minimal group effects. We robustly replicate minimal intergroup bias in mental representations of faces, evaluations, trait inferences, and resource allocations. At the same time, we show that seemingly arbitrary category labels can imply characteristics about groups that may influence responses in intergroup contexts. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Grid-like and distance codes for representing word meaning in the human brain

Relational information about items in memory is thought to be represented in our brain thanks to an internal comprehensive model, also referred to as a “cognitive map”. In the human neuroimaging literature, two signatures of bi-dimensional cognitive maps have been reported: the grid-like code and the distance-dependent code. While these kinds of representation were previously observed during spatial navigation and, more recently, during processing of perceptual stimuli, it is still an open question whether they also underlie the representation of the most basic items of language: words. Here we taught human participants the meaning of novel words as arbitrary labels for a set of audiovisual objects varying orthogonally in size and sound. The novel words were therefore conceivable as points in a navigable 2D map of meaning. While subjects performed a word comparison task, we recorded their brain activity using functional magnetic resonance imaging (fMRI). By applying a combination of representational similarity and fMRI-adaptation analyses, we found evidence of (i) a grid-like code, in the right postero-medial entorhinal cortex, representing the relative angular positions of words in the word space, and (ii) a distance-dependent code, in medial prefrontal, orbitofrontal, and mid-cingulate cortices, representing the Euclidean distance between words. Additionally, we found evidence that the brain also separately represents the single dimensions of word meaning: their implied size, encoded in visual areas, and their implied sound, in Heschl's gyrus/Insula. These results support the idea that the meaning of words, when they are organized along two dimensions, is represented in the human brain across multiple maps of different dimensionality. Significant statementHow do we represent the meaning of words and perform comparative judgements on them in our brain? According to influential theories, concepts are conceivable as points of an internal map (where distance represents similarity) that, as the physical space, can be mentally navigated. Here we use fMRI to show that when humans compare newly learnt words, they recruit a grid-like and a distance code, the same types of neural codes that, in mammals, represent relations between locations in the environment and support physical navigation between them.

Feature distribution-based label correlation in multi-label classification

In multi-label classification, multiple label variables in output space are equally important and can be predicted according to a common set of input variables. To improve the accuracy and efficiency of multi-label learner, measuring and utilizing label correlation is the core breakthrough. Extensive research on label correlation focuses on the co-occurrence or mutual exclusion frequency of label values in output space. In this paper, to handle the multi-label learning tasks, a novel method, named FL-MLC, is proposed by considering the influence of feature-label dependencies on inter-label correlations. In order to describe the intrinsic relationship between feature variable and label variable, the discriminant weight of any feature to label is first defined. Therefore, the concept of feature distribution for inputs on label is proposed to reflect the discriminant weights of features to the label. The corresponding calculation process is also designed based on multiple kernel learning and kernel alignment. Furthermore, the feature distributions on different labels are integrated into the feature distribution-based label correlation by using two different aggregation strategies. Obviously, arbitrary label variables with highly similar feature distributions have strong relevance. Thus, the feature distribution-based label correlation is applied to adjust the distance between the parameters for different labels in the predictive learner of FL-MLC method. Finally, the experimental results on twelve real-world datasets demonstrate that our methods achieves good effectiveness and versatility for multi-label classification.

An Ethnopragmatic Analysis of Death-Prevention Names in the Karanga Society of Zimbabwe

ABSTRACT This article interrogates death-prevention names among the Karanga people in Masvingo Province, Zimbabwe, from an ethnopragmatic perspective. In Karanga society, there is a strong belief that death-prevention names can counter the powers of supernatural forces that are responsible for the death of children. The use of such names is anchored in the Karanga belief that death is caused by external forces and can thus be prevented. The article argues that the bestowal of death-prevention names in the Karanga onomastic tradition is an emotional reaction to infant mortality and a device to facilitate recovery from fear, as the names are believed to prevent the death of the name-bearers. The article evinces that death triggers feelings of despair, helplessness and hopelessness driving name-givers to make entreaties to spiritual forces to spare the lives of the named children. It can also generate extreme frustration that can drive name-givers to express feelings of defiance and, in the case of death allegedly caused by witchcraft, threaten revenge through the names. The article uses data from four predominantly Karanga districts in Masvingo. The article concludes with the argument that death-prevention names are not arbitrary labels but are terse expressions of larger sociocultural meanings.

A graph-based semi-supervised approach to classification learning in digital geographies

As the distinction between online and physical spaces rapidly degrades, social media have now become an integral component of how many people's everyday experiences are mediated. As such, increasing interest has emerged in exploring how the content shared through those online platforms comes to contribute to the collaborative creation of places in physical space at the urban scale. Exploring digital geographies of social media data using methods such as qualitative coding (i.e., content labelling) is a flexible but complex task, commonly limited to small samples due to its impracticality over large datasets. In this paper, we propose a new tool for studies in digital geographies, bridging qualitative and quantitative approaches, able to learn a set of arbitrary labels (qualitative codes) on a small, manually-created sample and apply the same labels on a larger set. We introduce a semi-supervised, deep neural network approach to classify geo-located social media posts based on their textual and image content, as well as geographical and temporal aspects. Our innovative approach is rooted in our understanding of social media posts as augmentations of the time-space configurations that places are, and it comprises a stacked multi-modal autoencoder neural network to create joint representations of text and images, and a spatio-temporal graph convolution neural network for semi-supervised classification. The results presented in this paper show that our approach performs the classification of social media content with higher accuracy than traditional machine learning models as well as two state-of-art deep learning frameworks.

Consistent verbal labels promote odor category learning

Recent research shows that speakers of most languages find smells difficult to abstract and name. Can verbal labels enhance the human capacity to learn smell categories? Few studies have examined how verbal labeling might affect non-visual cognitive processes, and thus far very little is known about word-assisted odor category learning. To address these gaps, we tested whether different types of training change learning gains in odor categorization. After four intensive days of training to categorize odors that were co-presented with arbitrary verbal labels, people who learned odor categories with odor-label pairs that were more consistent were significantly more accurate than people with the same perceptual experience but who had odor-label pairs that were less consistent. Both groups' accuracy scores improved, but the learning curves differed. The context of consistent linguistic cuing supported an increase in correct responses from the third day of training. However, inconsistent linguistic cuing delayed the start of approximating to target odor categorization until after the fourth day. These results show that associations formed between odors and novel verbal labels facilitate the formation of odor categories. We interpret this as showing a causal link between language and olfactory perceptual processing in supporting categorization.

An Investigation into the Sociolinguistics of Asu Personal Names in Same, Tanzania

Many Ethnic Community Languages (ECLs) in Tanzania are demographically and socioculturally pressured mainly by Kiswahili and English to a lesser extent. The ECLs which were previously used in elementary education, local administration and religious activities currently do not have any place in any official domain and thus are limited to home and other few immediate domains. Due to this unequal coexistence of the languages, many ECLs are considered endangered, calling for efforts from stakeholders to prevent the death of such a precious cultural heritage. By documenting the social aspects of Asu personal names, this paper is a contribution to such initiatives like The Languages of Tanzania Project aiming at documenting Tanzanian ECLs in every possible area and means. The paper used participant observation, in-depth interview and self-intuition to investigate personal names of a Bantu speaking people called Vaasu (Asu) of Northern Tanzania, considering naming as an important aspect of the society. The paper looked at Asu names within the purview of linguistic anthropology considering names as not being arbitrary labels but sociocultural tags that have sociocultural functions and meanings. By using thematic analysis technique, the paper analysed and discussed the typology of the names including family names, circumstantial names, theophorous names, flora and fauna names, to mention but a few. The paper further examined the changing nature of Asu naming system and practice as dictated by cultural contact mainly with Swahili and Christian/ western culture. The paper eventually recommends for further investigation on issues surrounding naming practices and strategic measures to prevent this important African cultural resource.

The ethnopragmatics of Yoruba personal names: Language in the context of culture.

While the subject of Yoruba names has been significantly explored by previous studies, this paper discusses extensively the nature of such names from an ethnopragmatic framework, with the aim of explicating how Yoruba names are formed, their various cultural contexts as well as the significant functions they play in the Yoruba ethnolinguistic ecology. It identifies and categorizes personal names based on contexts such as family situation, circumstantces of birth, religious orientation, death situation and profession. This paper reinforces that names are not just arbitrary labels, but most notably, linguistic categories – lexical, phrasal or sentential – that have indexical relationship to sociocultural meanings and functions, places, time, people, and events.

AI Radar Sensor: Creating Radar Depth Sounder Images Based on Generative Adversarial Network.

Significant resources have been spent in collecting and storing large and heterogeneous radar datasets during expensive Arctic and Antarctic fieldwork. The vast majority of data available is unlabeled, and the labeling process is both time-consuming and expensive. One possible alternative to the labeling process is the use of synthetically generated data with artificial intelligence. Instead of labeling real images, we can generate synthetic data based on arbitrary labels. In this way, training data can be quickly augmented with additional images. In this research, we evaluated the performance of synthetically generated radar images based on modified cycle-consistent adversarial networks. We conducted several experiments to test the quality of the generated radar imagery. We also tested the quality of a state-of-the-art contour detection algorithm on synthetic data and different combinations of real and synthetic data. Our experiments show that synthetic radar images generated by generative adversarial network (GAN) can be used in combination with real images for data augmentation and training of deep neural networks. However, the synthetic images generated by GANs cannot be used solely for training a neural network (training on synthetic and testing on real) as they cannot simulate all of the radar characteristics such as noise or Doppler effects. To the best of our knowledge, this is the first work in creating radar sounder imagery based on generative adversarial network.

Automatic structure-based NMR methyl resonance assignment in large proteins

Isotopically labeled methyl groups provide NMR probes in large, otherwise deuterated proteins. However, the resonance assignment constitutes a bottleneck for broader applicability of methyl-based NMR. Here, we present the automated MethylFLYA method for the assignment of methyl groups that is based on methyl-methyl nuclear Overhauser effect spectroscopy (NOESY) peak lists. MethylFLYA is applied to five proteins (28–358 kDa) comprising a total of 708 isotope-labeled methyl groups, of which 612 contribute NOESY cross peaks. MethylFLYA confidently assigns 488 methyl groups, i.e. 80% of those with NOESY data. Of these, 459 agree with the reference, 6 were different, and 23 were without reference assignment. MethylFLYA assigns significantly more methyl groups than alternative algorithms, has an average error rate of 1%, modest runtimes of 0.4–1.2 h, and can handle arbitrary isotope labeling patterns and data from other types of NMR spectra.

Efficient Inference in Stochastic Block Models With Vertex Labels

We study the stochastic block model with two communities where vertices contain side information in the form of a vertex label. These vertex labels may have arbitrary label distributions, depending on the community memberships. We analyze a linearized version of the popular belief propagation algorithm. We show that this algorithm achieves the highest accuracy possible whenever a certain function of the network parameters has a unique fixed point. Whenever this function has multiple fixed points, the belief propagation algorithm may not perform optimally. We show that increasing the information in the vertex labels may reduce the number of fixed points and hence lead to optimality of belief propagation.