Shared Representation Research Articles

Negotiating change for gender equality: identifying leverages, overcoming barriers

Introduction. Institutional change through gender equality plans is today the dominant approach to promoting gender equality in higher education and research. Building on our experiences as “technical support partners” in several EU-funded projects, we reflect on how change is negotiated in a variety of contexts. Objectives. Theoretically, using Feminist Institutionalism and the Science and Technology Studies concept of the trading zone, our objective is to analyse institutional negotiations among various, differently positioned actors with diverse backgrounds, value systems and negotiating power. From a practice-oriented perspective, our aim is to demonstrate typical challenges, suggest pathways towards solutions, and identify specific negotiation skills which underscore the capacity-building needs of change agents. Methodology. For our analysis, we have selected eight information-rich case studies through purposive theory-based sampling, illustrating the different transactions in the trading zones, based on our prior knowledge of the circumstances. The methods we draw on are primarily participant observation and textual analysis of project documents. Results. The selected theoretical combination allows us to identify leverages, ways to overcome barriers and the required skills and competences. Specifically, we underscore the use of participatory and co-creation techniques, strategic framing, spotting and using windows of opportunity, and wide mobilisation of stakeholders. We highlight key features of the change process, including its processual and incremental nature, the need for constant negotiation and the capacity-building needs of change agents. Contribution. With this analysis, we contribute, firstly, to the understanding of organisational change by identifying concrete barriers and opportunities as well as considering the ways in which a shared representation of gender equality is developed. The second, theoretical contribution lies in combining Feminist Institutionalism and the concept of the trading zone, which allows us to bring to productive dialogue issues of power, processuality and the need to address both material and discursive enactments of change processes.

Majority Decision-Making Works Best Under Conditions of Leadership Ambiguity and Shared Task Representations.

The effectiveness of decision-making teams depends largely on their ability to integrate and make sense of information. Consequently, teams which more often use majority decision-making may make better quality decisions, but particularly so when they also have task representations which emphasize the elaboration of information relevant to the decision, in the absence of clear leadership. In the present study we propose that (a) majority decision-making will be more effective when task representations are shared, and that (b) this positive effect will be more pronounced when leadership ambiguity (i.e., team members’ perceptions of the absence of a clear leader) is high. These hypotheses were put to the test using a sample comprising 81 teams competing in a complex business simulation for seven weeks. As predicted, majority decision-making was more effective when task representations were shared, and this positive effect was more pronounced when there was leadership ambiguity. The findings extend and nuance earlier research on decision rules, the role of shared task representations, and leadership clarity.

Putting US First: How Outgroup Hostilities and Defense of the Status Quo Motivate White Evangelical Affect Toward Candidates in U.S. Elections 2004 to 2016

How do voters construct feelings toward inparty elites? More specifically, how do they do so when they lack a shared policy agenda or shared salient social identity with candidates beyond partisan identification? In this paper, I investigate this puzzle by developing a theory of Putting America First to explain white evangelical affect toward Republican presidential candidates from 2004 to 2016. Using ANES surveys from 2004 to 2016, I test the effectiveness of this model of candidate affect. I find that shared outgroup hostility, what I call Putting Us First, motivates positive affect for presidential candidates among white evangelicals, regardless of shared policy objectives and descriptive representation. Overtime, white evangelicals’ affect is driven by outgroup hostilities rather than Culture Wars values. Overall, this informs our understanding of voter affect toward candidates and the increasingly important role of social outgroup hostility in defense of the status quo.

Effects of dyadic discrepancy in awareness of mild cognitive impairment on changes in care partner’s burden over time

Objectives:This study explored interpersonal aspects of coping with mild cognitive impairment (MCI), focusing on how shared illness representations between persons with MCI (PwMCI) and their care partners (CPs) are associated with CP perceived burden. Method:Longitudinal data from a purposive sample of 69 MCI care dyads were analyzed using descriptive statistics, t-tests, intraclass correlations, and regression analyses. Results: CP perceived their PwMCI’s deficits in MCI-related functions as more negative than PwMCI rated themselves, but the extent of the discrepancy varied across cognitive and behavioral abilities. The discrepancy in dyadic appraisal of PwMCI functioning was associated significantly with CP burden. Conclusion: The findings highlight the value of shared PwMCI and CP perspectives on cognitive functioning for MCI management and psychological adjustment.

Examining the Relationship Between Speech Perception, Production Distinctness, and Production Variability.

Several studies have demonstrated that individuals’ ability to perceive a speech sound contrast is related to the production of that contrast in their native language. The theoretical account for this relationship is that speech perception and production have a shared multimodal representation in relevant sensory spaces (e.g., auditory and somatosensory domains). This gives rise to a prediction that individuals with more narrowly defined targets will produce greater separation between contrasting sounds, as well as lower variability in the production of each sound. However, empirical studies that tested this hypothesis, particularly with regard to variability, have reported mixed outcomes. The current study investigates the relationship between perceptual ability and production ability, focusing on the auditory domain. We examined whether individuals’ categorical labeling consistency for the American English /ε/–/æ/ contrast, measured using a perceptual identification task, is related to distance between the centroids of vowel categories in acoustic space (i.e., vowel contrast distance) and to two measures of production variability: the overall distribution of repeated tokens for the vowels (i.e., area of the ellipse) and the proportional within-trial decrease in variability as defined as the magnitude of self-correction to the initial acoustic variation of each token (i.e., centering ratio). No significant associations were found between categorical labeling consistency and vowel contrast distance, between categorical labeling consistency and area of the ellipse, or between categorical labeling consistency and centering ratio. These null results suggest that the perception-production relation may not be as robust as suggested by a widely adopted theoretical framing in terms of the size of auditory target regions. However, the present results may also be attributable to choices in implementation (e.g., the use of model talkers instead of continua derived from the participants’ own productions) that should be subject to further investigation.

Towards an automatic analysis of CHO-K1 suspension growth in microfluidic single-cell cultivation.

Innovative microfluidic systems carry the promise to greatly facilitate spatio-temporal analysis of single cells under well-defined environmental conditions, allowing novel insights into population heterogeneity and opening new opportunities for fundamental and applied biotechnology. Microfluidics experiments, however, are accompanied by vast amounts of data, such as time series of microscopic images, for which manual evaluation is infeasible due to the sheer number of samples. While classical image processing technologies do not lead to satisfactory results in this domain, modern deep-learning technologies, such as convolutional networks can be sufficiently versatile for diverse tasks, including automatic cell counting as well as the extraction of critical parameters, such as growth rate. However, for successful training, current supervised deep learning requires label information, such as the number or positions of cells for each image in a series; obtaining these annotations is very costly in this setting. We propose a novel machine-learning architecture together with a specialized training procedure, which allows us to infuse a deep neural network with human-powered abstraction on the level of data, leading to a high-performing regression model that requires only a very small amount of labeled data. Specifically, we train a generative model simultaneously on natural and synthetic data, so that it learns a shared representation, from which a target variable, such as the cell count, can be reliably estimated. The project is cross-platform, open-source and free (MIT licensed) software. We make the source code available at https://github.com/dstallmann/cell_cultivation_analysis; the dataset is available at https://pub.uni-bielefeld.de/record/2945513.

Multi-View Information-Bottleneck Representation Learning

In real-world applications, clustering or classification can usually be improved by fusing information from different views. Therefore, unsupervised representation learning on multi-view data becomes a compelling topic in machine learning. In this paper, we propose a novel and flexible unsupervised multi-view representation learning model termed Collaborative Multi-View Information Bottleneck Networks (CMIB-Nets), which comprehensively explores the common latent structure and the view-specific intrinsic information, and discards the superfluous information in the data significantly improving the generalization capability of the model. Specifically, our proposed model relies on the information bottleneck principle to integrate the shared representation among different views and the view-specific representation of each view, prompting the multi-view complete representation and flexibly balancing the complementarity and consistency among multiple views. We conduct extensive experiments (including clustering analysis, robustness experiment, and ablation study) on real-world datasets, which empirically show promising generalization ability and robustness compared to state-of-the-arts.

A novel gradient foster shared-representation convolutional network optimization for multi-modalities

Significant growth has been made with multi-modal data as its entrance in the field of deep learning; whereas, Convolutional Neural Network (CNN) provides sufficient training data to develop a representative encrusted image. Yet, the multi-modality approach in CNN affect the performance by slowly converge the variance along with high-dimensionality, heterogeneity and non-aconvex optimization problems. To abridge these issues, a novel Gradient Foster Shared-representation Convolutional Network (GFSCN) framework is proposed, which improve and optimize the performance interms of accuracy and dimensionality reduction. Initially, the framework incorporates a multiple scant weighted de-noising autoencoder to solve the heterogeneity problem and reduces the dimensionality of data by transforming shared feature representation. Consequently, the work integrated enhanced stochastic variance reduced ascension approach. This approach diminishes the non-convex optimization problem through integrating two gradients consuming mini-batches, which reduced the loss function thereby achieves faster convergence even with the usage of larger dataset. Thus, the proposed framework achieves better performance in terms of achieving utmost accuracy with faster convergence and reduced variance.

Shared Dorsal Periaqueductal Gray Activation Patterns during Exposure to Innate and Conditioned Threats.

The brainstem dorsal periaqueductal gray (dPAG) has been widely recognized as being a vital node orchestrating the responses to innate threats. Intriguingly, recent evidence also shows that the dPAG mediates defensive responses to fear conditioned contexts. However, it is unknown whether the dPAG displays independent or shared patterns of activation during exposure to innate and conditioned threats. It is also unclear how dPAG ensembles encode and predict diverse defensive behaviors. To address this question, we used miniaturized microscopes to obtain recordings of the same dPAG ensembles during exposure to a live predator and a fear conditioned context in male mice. dPAG ensembles encoded not only distance to threat, but also relevant features, such as predator speed and angular offset between mouse and threat. Furthermore, dPAG cells accurately encoded numerous defensive behaviors, including freezing, stretch-attend postures, and escape. Encoding of behaviors and of distance to threat occurred independently in dPAG cells. dPAG cells also displayed a shared representation to encode these behaviors and distance to threat across innate and conditioned threats. Last, we also show that escape could be predicted by dPAG activity several seconds in advance. Thus, dPAG activity dynamically tracks key kinematic and behavioral variables during exposure to threats, and exhibits similar patterns of activation during defensive behaviors elicited by innate or conditioned threats. These data indicate that a common pathway may be recruited by the dPAG during exposure to a wide variety of threat modalities.SIGNIFICANCE STATEMENT The dorsal periaqueductal gray (dPAG) is critical to generate defensive behaviors during encounters with threats of multiple modalities. Here we use longitudinal calcium transient recordings of dPAG ensembles in freely moving mice to show that this region uses shared patterns of activity to represent distance to an innate threat (a live predator) and a conditioned threat (a shock grid). We also show that dPAG neural activity can predict diverse defensive behaviors. These data indicate the dPAG uses conserved population-level activity patterns to encode and coordinate defensive behaviors during exposure to both innate and conditioned threats.

RPLS-Net: pulmonary lobe segmentation based on 3D fully convolutional networks and multi-task learning.

The robust and automatic segmentation of the pulmonary lobe is vital to surgical planning and regional image analysis of pulmonary related diseases in real-time Computer Aided Diagnosis systems. While a number of studies have examined this issue, the segmentation of unclear borders of the five lobes of the lung remains challenging because of incomplete fissures, the diversity of anatomical pulmonary information, and obstructive lesions caused by pulmonary diseases. This study proposes a model called Regularized Pulmonary Lobe Segmentation Network to accurately predict the lobes as well as the borders. First, a 3D fully convolutional network is constructed to extract contextual features from computed tomography images. Second, multi-task learning is employed to learn the segmentations of the lobes and the borders between them to train the neural network to better predict the borders via shared representation. Third, a 3D depth-wise separable de-convolution block is proposed for deep supervision to efficiently train the network. We also propose a hybrid loss function by combining cross-entropy loss with focal loss using adaptive parameters to focus on the tissues and the borders of the lobes. Experiments are conducted on a dataset annotated by experienced clinical radiologists. A 4-fold cross-validation result demonstrates that the proposed approach can achieve a mean dice coefficient of 0.9421 and average symmetric surface distance of 1.3546 mm, which is comparable to state of the art methods. The proposed approach has the capability to accurately segment voxels that are near the lung wall and fissure. In this paper, a 3D fully convolutional networks framework is proposed to segment pulmonary lobes in chest CT images accurately. Experimental results show the effectiveness of the proposed approach in segmenting the tissues as well as the borders of the lobes.

Knowledge order in an online knowledge community: Group heterogeneity and two paths mediated by group interaction

AbstractKnowledge order in an online knowledge community (OKC) refers to a consensual version of collective knowledge in the creation of shared knowledge representation. Much previous research has been conducted in the context of the ordered structure of objective knowledge systems, but this does little to explain the microlevel order of knowledge after users contribute knowledge and achieve consensus through online interactions in OKC. Based on interactive team cognition theory and the stigmergy coordination mechanism, our research aims to investigate how knowledge and experience heterogeneity affect knowledge order effectiveness and efficiency through collaborative and communicative interaction. To test our hypotheses, we randomly collected the records of 250 articles from the English version of Wikipedia. Partial least squares structural equation modeling indicated that OKC favoring online collective knowledge order by limiting communicative interaction, as collaborative interaction is very effective in achieving knowledge order and in achieving it in a fast way. From our findings, scholars and practitioners are advised to pay attention to online knowledge order in the management and design of OKC.

PTP: Parallelized Tracking and Prediction With Graph Neural Networks and Diversity Sampling

Multi-object tracking (MOT) and trajectory prediction are two critical components in modern 3D perception systems that require accurate modeling of multi-agent interaction. We hypothesize that it is beneficial to unify both tasks under one framework in order to learn a shared feature representation of agent interaction. Furthermore, instead of performing tracking and prediction sequentially which can propagate errors from tracking to prediction, we propose a parallelized framework to mitigate the issue. Also, our parallel track-forecast framework incorporates two additional novel computational units. First, we use a feature interaction technique by introducing Graph Neural Networks (GNNs) to capture the way in which agents interact with one another. The GNN is able to improve discriminative feature learning for MOT association and provide socially-aware contexts for trajectory prediction. Second, we use a diversity sampling function to improve the quality and diversity of our forecasted trajectories. The learned sampling function is trained to efficiently extract a variety of outcomes from a generative trajectory distribution and helps avoid the problem of generating duplicate trajectory samples. We evaluate on KITTI and nuScenes datasets showing that our method with socially-aware feature learning and diversity sampling achieves new state-of-the-art performance on 3D MOT and trajectory prediction. Project website is: http://www.xinshuoweng.com/projects/PTP.

Neural substrates for sharing intention in action during face-to-face imitation

Face-to-face imitation is a unique social interaction wherein a shared action is executed based on the feedback of the partner. Imitation by the partner is the feedback to the imitatee's action, resulting in sharing actions. The neural mechanisms of the shared representation of action during face-to-face imitation, the core of inter-subjectivity, are not well-known. Here, based on the predictive coding account, we hypothesized that the pair-specific forward internal model is the shared representation of action which is represented by the inter-individual synchronization of some portion of the mirror neuron system. Hyperscanning functional magnetic resonance imaging was conducted during face-to-face interaction in 16 pairs of participants who completed an immediate imitation task of facial expressions. Paired participants were alternately assigned to either an imitator or an imitatee who was prompted to express a happy, sad, or non-emotional face. While neural activation elicited by imitating and being imitated were distinct with little overlap, on-line imitative interaction enhanced inter-brain synchronization in the right inferior parietal lobule that correlated with the similarity in facial movement kinematic profile. This finding indicates a critical role of the right inferior parietal lobule in sharing representation of action as a pair-specific forward internal model through imitative interaction.

Body parts relevance learning via expectation–maximization for human pose estimation

Recently, most existing human pose estimation methods fuse multi-stage convolutional modules to learn a shared feature representation. In this paper, we propose a expectation–maximization (EM) mapping-based network to learn specific related body parts for human pose estimation, named EMposeNet. It maps specific feature of related parts from the original fully shared feature space. From the perspective of multi-task learning, we can regard the task of human pose estimation as a homogeneous multi-task learning. Sharing features among related tasks can result in a more compact model and better generalization ability. However, sharing features for those unrelated or weakly related tasks will deteriorate the estimation performance. Our proposed method aims at performing EM algorithm to learn the related body part, where the predicted keypoint heatmap is potentially more accurate and spatially more precise. We conduct extensive experiments on two benchmark datasets, including the MSCOCO keypoint detection dataset and the MPII human pose dataset, to empirically demonstrate the validity of the proposed method. The results on such two benchmark datasets show that the proposed approach achieves a competitive performance.

Relation-Induced Multi-Modal Shared Representation Learning for Alzheimer's Disease Diagnosis.

The fusion of multi-modal data (e.g., magnetic resonance imaging (MRI) and positron emission tomography (PET)) has been prevalent for accurate identification of Alzheimer's disease (AD) by providing complementary structural and functional information. However, most of the existing methods simply concatenate multi-modal features in the original space and ignore their underlying associations which may provide more discriminative characteristics for AD identification. Meanwhile, how to overcome the overfitting issue caused by high-dimensional multi-modal data remains appealing. To this end, we propose a relation-induced multi-modal shared representation learning method for AD diagnosis. The proposed method integrates representation learning, dimension reduction, and classifier modeling into a unified framework. Specifically, the framework first obtains multi-modal shared representations by learning a bi-directional mapping between original space and shared space. Within this shared space, we utilize several relational regularizers (including feature-feature, feature-label, and sample-sample regularizers) and auxiliary regularizers to encourage learning underlying associations inherent in multi-modal data and alleviate overfitting, respectively. Next, we project the shared representations into the target space for AD diagnosis. To validate the effectiveness of our proposed approach, we conduct extensive experiments on two independent datasets (i.e., ADNI-1 and ADNI-2), and the experimental results demonstrate that our proposed method outperforms several state-of-the-art methods.

Interpersonal Motor Interactions Shape Multisensory Representations of the Peripersonal Space.

This perspective review focuses on the proposal that predictive multisensory integration occurring in one’s peripersonal space (PPS) supports individuals’ ability to efficiently interact with others, and that integrating sensorimotor signals from the interacting partners leads to the emergence of a shared representation of the PPS. To support this proposal, we first introduce the features of body and PPS representations that are relevant for interpersonal motor interactions. Then, we highlight the role of action planning and execution on the dynamic expansion of the PPS. We continue by presenting evidence of PPS modulations after tool use and review studies suggesting that PPS expansions may be accounted for by Bayesian sensory filtering through predictive coding. In the central section, we describe how this conceptual framework can be used to explain the mechanisms through which the PPS may be modulated by the actions of our interaction partner, in order to facilitate interpersonal coordination. Last, we discuss how this proposal may support recent evidence concerning PPS rigidity in Autism Spectrum Disorder (ASD) and its possible relationship with ASD individuals’ difficulties during interpersonal coordination. Future studies will need to clarify the mechanisms and neural underpinning of these dynamic, interpersonal modulations of the PPS.

MustGAN: multi-stream Generative Adversarial Networks for MR Image Synthesis.

Multi-contrast MRI protocols increase the level of morphological information available for diagnosis. Yet, the number and quality of contrasts are limited in practice by various factors including scan time and patient motion. Synthesis of missing or corrupted contrasts from other high-quality ones can alleviate this limitation. When a single target contrast is of interest, common approaches for multi-contrast MRI involve either one-to-one or many-to-one synthesis methods depending on their input. One-to-one methods take as input a single source contrast, and they learn a latent representation sensitive to unique features of the source. Meanwhile, many-to-one methods receive multiple distinct sources, and they learn a shared latent representation more sensitive to common features across sources. For enhanced image synthesis, we propose a multi-stream approach that aggregates information across multiple source images via a mixture of multiple one-to-one streams and a joint many-to-one stream. The complementary feature maps generated in the one-to-one streams and the shared feature maps generated in the many-to-one stream are combined with a fusion block. The location of the fusion block is adaptively modified to maximize task-specific performance. Quantitative and radiological assessments on T1,- T2-, PD-weighted, and FLAIR images clearly demonstrate the superior performance of the proposed method compared to previous state-of-the-art one-to-one and many-to-one methods.

Knowledge-aware multi-center clinical dataset adaptation: Problem, method, and application

Adaptable utilization of clinical data collected from multiple centers, prompted by the need to overcome the shifts between the dataset distributions, and exploit these different datasets for potential clinical applications, has received significant attention in recent years. In this study, we propose a novel approach to this task by infusing an external knowledge graph (KG) into multi-center clinical data mining. Specifically, we propose an adversarial learning model to capture shared patient feature representations from multi-center heterogeneous clinical datasets, and employ an external KG to enrich the semantics of the patient sample by providing both clinical center-specific and center-general knowledge features, which are trained with a graph convolutional autoencoder. We evaluate the proposed model on a real clinical dataset extracted from the general cardiology wards of a Chinese hospital and a well-known public clinical dataset (MIMIC III, pertaining to ICU clinical settings) for the task of predicting acute kidney injury in patients with heart failure. The achieved experimental results demonstrate the efficacy of our proposed model.

MTTLADE: A multi-task transfer learning-based method for adverse drug events extraction

Extracting mentions of Adverse Drug Events (ADEs) and the potential relationships among them from clinical textual data remains challenging tasks due to the following issues: (1) many ADEs mentions have multiple relations, also known as the multi-head issue, and (2) many ADEs relations contain discontinuous mentions. To deal with these problems, in this paper, we propose a Multi-Task Transfer Learning-based method for ADEs extraction, called MTTLADE. Firstly, the MTTLADE system converts the ADEs extraction task to a dual-task sequence labelling which includes ADEs source mention extraction (ADE-SE) and ADEs attribute-relation extraction (ADE-Att-RE) tasks. The ADE-SE task aims at extracting the source mentions that are likely related to at least one relation, while the ADE-Att-RE task consists in linking the previously identified source mentions to their target attributes and relation types by adopting a unified sequence labelling. Then, it uses the multi-task transfer learning (MTTL) based approach to process the two proposed tasks simultaneously. The MTTL adopts a shared representation obtained from the pre-trained language model learned through transformer architecture and ends up with task-specific fine-tuning. This allows the MTTLADE system to yield more generalized representation across the tasks. Finally, MTTLADE produces sequences for each task from the generated model so as to extract ADEs mentions and relations. Experimental evaluations conducted on two datasets provided by the TAC 2017 and n2c2 2018 shared tasks show the effectiveness and generalizability of MTTLADE. The proposed MTTLADE system significantly outperforms the state-of-the-art ones on both datasets. The results also show that combining transfer and multi-task learning makes MTTLADE more effective for solving the multi-head issue and extracting intricate ADEs.

Shared syntactic representations in bilinguals

Abstract The current study explores the nature of constituent-structure-independent structural priming across the two languages of bilinguals. Specifically, this study tests whether such cross-linguistic priming involves the priming of functional-level syntactic representations shared between the languages, which can be distinguished from the priming of mainly non-syntactic information (e.g., information structure, thematic-role order). Critical prime sentences consisted of Cantonese actives in the Object-Subject-Verb (OSV) order and passives where the patient was grammatically topicalized with the same topic particle. Target responses were produced in English actives or passives. The results show that robust priming from Cantonese Topic-Passives to English passives occurred, but no cross-linguistic priming was observed for Cantonese Topic-OSV active primes. The Topic-OSV active and Topic-Passive constructions share information structure, and are formed in different constituent structures from English actives and passives. Therefore, the robust cross-linguistic passive priming by Topic-Passive primes should in large part be ascribed to functional-level syntactic representations of passive constructions shared by Cantonese and English.