Task Representation Research Articles

Inhibition during task switching is affected by the number of competing tasks.

Inhibition during task switching is assumed to be indexed by n - 2 repetition costs-that is, performance costs when the task in the current trial equals the task in trial n - 2 (sequences of type ABA) compared with two consecutive switches to another task each (sequences CBA). The present study examined effects of a short-term reduction of the number of candidate tasks on these costs. For this purpose, a variant of the task switching paradigm was used in which in half of the trials, a precue that preceded the task cue allowed for a short-term reduction of the number of candidate tasks. In Experiment 1, one out of three tasks could be excluded. In Experiment 2, one or two out of four tasks could be excluded. Experiment 3 served as control condition using the standard cueing paradigm. Significant n - 2 repetition costs were present with three candidate tasks. In contrast, no costs were visible when the number of candidate tasks was reduced to two. This result is interpreted in terms of a task selection mechanism based on antagonistic constraints among task representations, which operates on a rather superficial level when switching among only two tasks, thereby reducing the need for inhibition.

A Multi-Story Expandable Systematic Hierarchical Construction Information Classification System for Implementing Information Processing in Highway Construction

In the field of infrastructure construction, progress in digital transformation remains limited; this is particularly true in road construction, an infrastructure facility involving design, construction, and operation stages. Many construction subjects are involved at each stage of this cycle, generating substantial construction information. To drive the digital transformation of the construction industry, a construction information classification system is necessary for the development of a systematic construction information model. This study focuses on categorizing construction information into objects and activities, defining unit work by combining these two categories, and allowing for the flexible processing of construction information. A construction information classification system was developed, representing the evolving construction-related methodological information throughout the project’s lifecycle. Applying this multi-story expandable systematic hierarchical system to a highway project demonstrates the representation of key tasks in each construction phase, enabling the future stage-specific expansion of construction information for individual tasks. The proposed system could advance the digital transformation in infrastructure construction.

RepCo: Replenish sample views with better consistency for contrastive learning

Contrastive learning methods aim to learn shared representations by minimizing distances between positive pairs, and maximizing distances between negative pairs in the embedding space. To achieve better performance of contrastive learning, one of the key problems is to design appropriate sample pairs. In most previous works, random cropping on the input image is utilized to obtain two views as positive pairs. However, such strategies lead to suboptimal performance since the sampled crops may have inconsistent semantic information, which consequently degrades the quality of contrastive views. To address this limitation, we explore to replenish sample views with better consistency of the image and propose a novel self-supervised learning (SSL) framework RepCo. Instead of searching for semantically consistent patches between two different views, we select patches on the same image as the replenishment of positive/negative pairs, encourage patches that are similar but come from different positions as positive pairs, and force patches that are dissimilar but come from adjacent positions to have different representations, i.e. construct negative pairs to enrich the learned representations. Our method effectively generates high-quality contrastive views, explores the untapped semantic consistency on images, and provides more informative representations for downstream tasks. Experiments on adequate downstream tasks have shown that, our approach achieves ＋2.1 AP50 (COCO pre-trained) and ＋1.6 AP50 (ImageNet pre-trained) gains on Pascal VOC object detection, ＋2.3 mIoU gains on Cityscapes semantic segmentation, respectively.

Task-dependent optimal representations for cerebellar learning.

The cerebellar granule cell layer has inspired numerous theoretical models of neural representations that support learned behaviors, beginning with the work of Marr and Albus. In these models, granule cells form a sparse, combinatorial encoding of diverse sensorimotor inputs. Such sparse representations are optimal for learning to discriminate random stimuli. However, recent observations of dense, low-dimensional activity across granule cells have called into question the role of sparse coding in these neurons. Here, we generalize theories of cerebellar learning to determine the optimal granule cell representation for tasks beyond random stimulus discrimination, including continuous input-output transformations as required for smooth motor control. We show that for such tasks, the optimal granule cell representation is substantially denser than predicted by classical theories. Our results provide a general theory of learning in cerebellum-like systems and suggest that optimal cerebellar representations are task-dependent.

Numerical and Visual Representations of Uncertainty Lead to Different Patterns of Decision Making.

Although visualizations are a useful tool for helping people to understand information, they can also have unintended effects on human cognition. This is especially true for uncertain information, which is difficult for people to understand. Prior work has found that different methods of visualizing uncertain information can produce different patterns of decision making from users. However, uncertainty can also be represented via text or numerical information, and few studies have systematically compared these types of representations to visualizations of uncertainty. We present two experiments that compared visual representations of risk (icon arrays) to numerical representations (natural frequencies) in a wildfire evacuation task. Like prior studies, we found that different types of visual cues led to different patterns of decision making. In addition, our comparison of visual and numerical representations of risk found that people were more likely to evacuate when they saw visualizations than when they saw numerical representations. These experiments reinforce the idea that design choices are not neutral: seemingly minor differences in how information is represented can have important impacts on human risk perception and decision making.

Environmental Intervention Evaluation to Support Motor Planning of Wheelchair Movements: A Pilot Study

Environmental interventions provide a promising opportunity to support inclusive mobility in our rapidly changing society and workforce. Concurrently, the trends for aging into mobility disability, working into older age, and health and economic burdens of unpaid caregiving continue to rise. Hence, providing equitable mobility for all is an urgent need. In the context of manual wheelchair control, a novel method investigating the impact of environmental interventions on individual’s internal representation of motor actions is proposed. In addition to measuring objective and subjective performance metrics, we propose that an understanding of how internal representations of tasks are influenced by user perception of the environment has promising potential towards informing effective and inclusive task-environment interactions. The application explored in this pilot study aims to facilitate mobility.

Adversarial Cluster-Level and Global-Level Graph Contrastive Learning for node representation

Graph contrastive learning aims to learn informative and discriminative node representations for downstream tasks by maximizing the mutual information between representations of different augmentation views of the same node. However, according to the multi-view information bottleneck principle, redundant information in learned representations can negatively impact the performance of downstream tasks. To avoid this issue, we propose Adversarial Cluster-Level and Global-Level Graph Contrastive Learning (ACG-GCL) for learning minimal sufficient node representations. ACG-GCL is optimized alternately under an adversarial learning framework with a min–max objective. At the min step, ACG-GCL eliminates redundant information in both the graph structure and feature content of nodes while producing a new graph that provides multi-view information. At the max step, ACG-GCL seeks to preserve shared task-relevant information in the learned representation by maximizing the mutual information of node-cluster level and node-global level representations. We demonstrate the effectiveness of the proposed method on the tasks of node classification and node clustering tasks. Code is available at https://github.com/tangq123/ACG-GCL.

Images as Metadata: A New Perspective for Describing Research Data

Indispensable in many contexts, images are fundamental in the tasks of representation and transmission of information. In the scientific context, images can be tools for researchers seeking to see their data properly managed. Research data management guides in this direction as it determines necessary phases in the life cycle of projects. The description phase is fundamental as it is an essential means for data context, safeguarding, and reuse. The description often occurs through metadata models composed of descriptors capable of attributing context. However, there is one common aspect: the values associated with these descriptors are always textual or numeric. Through studies and work developed over the last few years, we propose a new approach to description, where images can have a preponderant role in the description of data, assuming the role of metadata. We present several pieces of evidence, point out their challenges and determine the opportunities this new perspective can have in the research. Images have specific characteristics that can be leveraged in improving data description. Historical evidence establish that images have always been used and produced in research, yet their representational ability has never been harnessed to describe data and give more context to the scientific process.

EventKGE: Event knowledge graph embedding with event causal transfer

Traditional knowledge graph embedding (KGE) aims to map entities and relations into continuous space vectors to provide high-quality data feature representation for downstream tasks. However, relations in most KGs often only reflect connections between static entities, but cannot represent dynamic activities and state changes of related entities, which makes the KGE models unable to effectively learn rich and comprehensive entity representation. In this paper, we verify the importance of embedding event knowledge in KG representation learning, and propose a novel event KGE model based on event causal transfer (EventKGE), which can effectively maintain the semantic information of events, entities, and relations in the event KG. First, we define a six tuple-based event representation model consisting of event trigger words, event arguments and event description text, which can effectively express events in a structured form. Second, for a given event KG, the event nodes and entity nodes in the KG are integrated through the constructed heterogeneous graph. Meanwhile, in the heterogeneous graph, the event nodes and entity nodes are connected through the event arguments type, and the event nodes are connected through the causal relationship. Finally, an information transfer method based on an attention network is designed, which is used for the relations between events, events and entities, and entities and entities, to integrate event information into KGE. Comprehensive experiments on the ACE2005 corpus and CEC2.0 dataset verify that the event KGE model we designed is efficient and stable in multiple downstream tasks.

A Spatio-Temporal Task Allocation Model in Mobile Crowdsensing Based on Knowledge Graph

With the increasing popularity of wireless networks and the development of smart cities, the Mobile Crowdsourcing System (MCS) has emerged as a framework for automatically assigning spatiotemporal tasks to workers. The study of mobile crowdsourcing makes a valuable research contribution to community service and urban route planning. However, previous algorithms have faced challenges in effectively addressing task allocation issues with massive spatial data. In this paper, we propose a novel solution to the spatiotemporal task allocation problem using a knowledge graph. Firstly, we construct a robust spatiotemporal knowledge graph (STKG) and employ a knowledge graph embedding algorithm to learn the representations of nodes and edges. Next, we utilize these representations to build a task transition graph, which is a weighted and learning-based graph that highlights important neighbors for each task. We then apply a simplified Graph Convolutional Network (GCN) and an RNN-based model to enhance task representations and capture sequential transition patterns on the task transition graph. Furthermore, we design a similarity function to facilitate personalized task allocation. Through experimental results, we demonstrate that our solution achieves higher accuracy compared to existing approaches when tested on three real datasets. These research findings are significant as they contribute to an 18.01% improvement in spatiotemporal task allocation accuracy.

Investigation of generalisation processes of secondary school students using multiple representations in a pattern task

The aim of this study is to show how the GeoGebra software-supported collaborative learning environment contributes to a group of students using multiple representations and reaching generalisations while solving problems collectively. Blending individual and collaborative learning activities were carried out with a learning method called ACODESA (from French: Apprentissage Collaboratif, Débat Scientifique, Auto-réflexion). Within this context, four 8th-grade students participating in the study were asked to solve the pattern problem by using the GeoGebra software and following the stages of the ACODESA method. The data were collected via a pattern problem developed by the researcher, students’ audio records and screen recordings, GeoGebra files and written productions. The data were analysed using Duval’s theory of registers of semiotic representation and Toulmin’s model. The Toulmin model was used to analyse students’ argument structures, that is, to determine their reasoning structures. The data analysis revealed that a GeoGebra software-supported collaborative learning environment supported secondary school students in using multiple representations, making algebraic reasoning and generalisation. The discussions in this process contributed to the students’ transformation of representations and the evolution of their representations. Providing students with the opportunity to express and examine the pattern with multiple representations, GeoGebra software helped students to obtain data, test their conjectures and thus reach generalisation.

Content preserving image translation with texture co-occurrence and spatial self-similarity for texture debiasing and domain adaptation

Models trained on datasets with texture bias usually perform poorly on out-of-distribution samples since biased representations are embedded into the model. Recently, various image translation and debiasing methods have attempted to disentangle texture biased representations for downstream tasks, but accurately discarding biased features without altering other relevant information is still challenging. In this paper, we propose a novel framework that leverages image translation to generate additional training images using the content of a source image and the texture of a target image with a different bias property to explicitly mitigate texture bias when training a model on a target task. Our model ensures texture similarity between the target and generated images via a texture co-occurrence loss while preserving content details from source images with a spatial self-similarity loss. Both the generated and original training images are combined to train improved classification or segmentation models robust to inconsistent texture bias. Evaluation on five classification- and two segmentation-datasets with known texture biases demonstrates the utility of our method, and reports significant improvements over recent state-of-the-art methods in all cases.

Clustered Task-Aware Meta-Learning by Learning From Learning Paths.

To enable effective learning of new tasks with only a few examples, meta-learning acquires common knowledge from the existing tasks with a globally shared meta-learner. To further address the problem of task heterogeneity, recent developments balance between customization and generalization by incorporating task clustering to generate task-aware modulation to be applied to the global meta-learner. However, these methods learn task representation mostly from the features ofinput data, while the task-specific optimization process with respect to the base-learner is often neglected. In this work, we propose a Clustered Task-Aware Meta-Learning (CTML) framework with task representation learned from both features and learning paths. We first conduct rehearsed task learning from the common initialization, and collect a set of geometric quantities that adequately describes this learning path. By inputting this set of values into a meta path learner, we automatically abstract path representation optimized for downstream clustering and modulation. Aggregating the path and feature representations results in an improved task representation. To further improve inference efficiency, we devise a shortcut tunnel to bypass the rehearsed learning process at a meta-testing time. Extensive experiments on two real-world application domains: few-shot image classification and cold-start recommendation demonstrate the superiority of CTML compared to state-of-the-art methods. We provide our code at https://github.com/didiya0825.

Rethinking Semantic Image Compression: Scalable Representation With Cross-Modality Transfer

This article proposes the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">scalable cross-modality compression</i> (SCMC) paradigm, in which the image compression problem is further cast into a representation task by hierarchically sketching the image with different modalities. Herein, we adopt the conceptual organization philosophy to model the overwhelmingly complicated visual patterns, based upon the semantic, structure, and signal level representation accounting for different tasks. The SCMC paradigm that incorporates the representation at different granularities supports diverse application scenarios, such as high-level semantic communication and low-level image reconstruction. The decoder, which enables the recovery of the visual information, benefits from the scalable coding based upon the semantic, structure, and signal layers. Qualitative and quantitative results demonstrate that the SCMC can convey accurate semantic and perceptual information of images, especially at low bitrates, and promising rate-distortion performance has been achieved compared to state-of-the-art methods. The code will be available online https://github.com/ppingzhang/SCMC.

Disentangled Graph Contrastive Learning With Independence Promotion

Self-supervised learning for graph neural networks has attracted considerable attention and shows notable successes in graph representation learning. However, the formation of a real-world graph typically arises from highly complex interactions of many latent factors. The existing self-supervised learning methods for GNNs are inherently holistic and neglect the entanglement of the latent factors, resulting in suboptimal learned representations for downstream tasks and difficult to be interpreted. Learning disentangled graph representations with self-supervised learning poses great challenges and remains largely ignored by the existing literature. In this paper, we introduce Independence Promoted Disentangled Graph Contrastive Learning ( <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">IDGCL</b> ) method, which can learn disentangled graph-level representations with self-supervision. In particular, we first identify the latent factors of the input graph and derive its factorized representations. Then we propose a factor-wise discrimination objective in a contrastive learning manner, which can force the factorized representations to independently reflect the expressive information from different latent factors. To further promote the independence between the representations, we employ the Hilbert-Schmidt Independence Criterion to eliminate the dependence among different representations, which is effectively integrated into the self-supervised framework as a regularizer. Extensive experiments on synthetic and real-world datasets demonstrate the superiority of our method against several state-of-the-art baselines.

Knowledge-Based Reasoning Network for Relation Detection.

With the rapid growth of large-scale knowledge bases (KBs), knowledge base question answering (KBQA) has attracted increasing attention recently. Relation detection plays an important role in the KBQA system, which finds a compatible answer by analyzing the semantics of questions and querying and reasoning with multiple KB triples. Significant progress has been made by deep neural networks. However, existing methods often concern on detecting single-hop relation without path reasoning, and a few of these methods exploit the multihop relation reasoning, which involves the answer reasoning from the noisy and abundant relational paths in the KB. Meanwhile, the relatedness between question and answer candidates has received little attention and remains unsolved. This article proposes a novel knowledge-based reasoning network (KRN) for relation detection, including both single-hop relation and multihop relation. To address the semantic gap problem in question-answer interaction, we first learn attentive question representations according to the influence of answer aspects. Then, we learn the single-hop relation sequence through different levels of abstraction and adopt the KB entity and structure information to denoise the multihop relation detection task. Finally, we adopt a Siamese network to measure the similarity between question representation and relation representation for both single-hop and multihop relation KBQA tasks. We conduct experiments on two well-known benchmarks, SimpleQuestions and WebQSP, and the results show the superiority of our approach over the state-of-the-art models for both single-hop and multihop relation detection. Our model also achieves a significant improvement over existing methods on KBQA end task. Further analysis demonstrates the robustness and the applicability of the proposed approach.

A variational autoencoder provides novel, data-driven features that explain functional brain representations in a naturalistic navigation task

A variational autoencoder provides novel, data-driven features that explain functional brain representations in a naturalistic navigation task

온라인 읽기 쓰기 수업에 RESOLV(Reader’s Representation of Reading Contexts and Tasks)모형의 적용 가능성 탐색 연구

In this study, we sought to find implications from the RESOLV model (Reader's Representation of Reading Contexts and Tasks) for designing online multi-text literacy lessons that engage students and improve their literacy skills. To understand the RESOLV model, we first examined its components, the Context Model and the Task Model, together. Based on our understanding of the RESOLV model, we proposed three things to consider when organizing online multidisciplinary literacy classes. First, we proposed how to organize tasks in online literacy classes based on the Context Model. Second, we proposed how to organize educational content in online literacy classes based on the Task Model. Third, we proposed a teaching and learning method that can promote collaborative interaction through team discussions with fellow learners as a teaching method for online reading and writing classes based on the integration of the situation model and the task model.

Abstraction of Reward Context Facilitates Relative Reward Coding in Neural Populations of the Macaque Anterior Cingulate Cortex.

The anterior cingulate cortex (ACC) is believed to be involved in many cognitive processes, including linking goals to actions and tracking decision-relevant contextual information. ACC neurons robustly encode expected outcomes, but how this relates to putative functions of ACC remains unknown. Here, we approach this question from the perspective of population codes by analyzing neural spiking data in the ventral and dorsal banks of the ACC in two male monkeys trained to perform a stimulus-motor mapping task to earn rewards or avoid losses. We found that neural populations favor a low dimensional representational geometry that emphasizes the valence of potential outcomes while also facilitating the independent, abstract representation of multiple task-relevant variables. Valence encoding persisted throughout the trial, and realized outcomes were primarily encoded in a relative sense, such that cue valence acted as a context for outcome encoding. This suggests that the population coding we observe could be a mechanism that allows feedback to be interpreted in a context-dependent manner. Together, our results point to a prominent role for ACC in context setting and relative interpretation of outcomes, facilitated by abstract, or untangled, representations of task variables.SIGNIFICANCE STATEMENT The ability to interpret events in light of the current context is a critical facet of higher-order cognition. The ACC is suggested to be important for tracking contextual information, whereas alternate views hold that its function is more related to the motor system and linking goals to appropriate actions. We evaluated these possibilities by analyzing geometric properties of neural population activity in monkey ACC when contexts were determined by the valence of potential outcomes and found that this information was represented as a dominant, abstract concept. Ensuing outcomes were then coded relative to these contexts, suggesting an important role for these representations in context-dependent evaluation. Such mechanisms may be critical for the abstract reasoning and generalization characteristic of biological intelligence.

Self-Supervised Electroencephalogram Representation Learning for Automatic Sleep Staging: Model Development and Evaluation Study.

Deep learning models have shown great success in automating tasks in sleep medicine by learning from carefully annotated electroencephalogram (EEG) data. However, effectively using a large amount of raw EEG data remains a challenge. In this study, we aim to learn robust vector representations from massive unlabeled EEG signals, such that the learned vectorized features (1) are expressive enough to replace the raw signals in the sleep staging task, and (2) provide better predictive performance than supervised models in scenarios involving fewer labels and noisy samples. We propose a self-supervised model, Contrast with the World Representation (ContraWR), for EEG signal representation learning. Unlike previous models that use a set of negative samples, our model uses global statistics (ie, the average representation) from the data set to distinguish signals associated with different sleep stages. The ContraWR model is evaluated on 3 real-world EEG data sets that include both settings: at-home and in-laboratory EEG recording. ContraWR outperforms 4 recently reported self-supervised learning methods on the sleep staging task across 3 large EEG data sets. ContraWR also supersedes supervised learning when fewer training labels are available (eg, 4% accuracy improvement when less than 2% of data are labeled on the Sleep EDF data set). Moreover, the model provides informative, representative feature structures in 2D projection. We show that ContraWR is robust to noise and can provide high-quality EEG representations for downstream prediction tasks. The proposed model can be generalized to other unsupervised physiological signal learning tasks. Future directions include exploring task-specific data augmentations and combining self-supervised methods with supervised methods, building upon the initial success of self-supervised learning reported in this study.