Performance Of Multiple Tasks Research Articles

Age-related enhancement in visuomotor learning by a dual-task

Many daily activities require performance of multiple tasks integrating cognitive and motor processes. While the fact that both processes go through deterioration and changes with aging has been generally accepted, not much is known about how aging interacts with stages of motor skill acquisition under a cognitively demanding situation. To address this question, we combined a visuomotor adaptation task with a secondary cognitive task. We made two primary findings beyond the expected age-related performance deterioration. First, while young adults showed classical dual-task cost in the early motor learning phase dominated by explicit processes, older adults instead strikingly displayed enhanced performance in the later stage, dominated by implicit processes. For older adults, the secondary task may have facilitated a shift to their relatively intact implicit learning processes that reduced reliance on their already-deficient explicit processes during visuomotor adaptation. Second, we demonstrated that consistently performing the secondary task in learning and re-learning phases can operate as an internal task-context and facilitate visuomotor memory retrieval later regardless of age groups. Therefore, our study demonstrated age-related similarities and differences in integrating concurrent cognitive load with motor skill acquisition which, may in turn, contributes to the understanding of a shift in balance across multiple systems.

Food Insecurity across the Life-Course and Cognitive Function among Older Mexican Adults.

Background: Food insecurity remains a global public health problem. Experiencing food insecurity is related to poorer cognitive function among older adults. However, few studies have examined how food insecurity, experienced over the life-course, relates to cognitive function among older adults in Mexico. Methods: Data came from the 2015 Mexican Health and Aging Study (n = 11,507 adults aged 50 and over). Early- and late-life food insecurity were ascertained by self-report. We evaluated how both measures of food insecurity related to the performance of multiple cognitive tasks (Verbal Learning, Verbal Recall, Visual Scanning, and Verbal Fluency), while controlling for key health and sociodemographic confounders using linear regression. Results: In descriptive analyses, respondents who experienced food insecurity in either early or late life performed significantly worse on all cognitive tasks when compared to the food secure. In models adjusted for health and sociodemographic confounders, early-life food insecurity predicted worse Verbal Learning performance and late-life food insecurity was associated with poorer Visual Scanning performance. Conclusions: Food insecurity was related to poorer cognitive function in a nationally representative sample of older adults in Mexico. However, results suggested that the significance of effects depended on cognitive task and when in the life-course food insecurity was experienced.

A multi-modal fusion framework based on multi-task correlation learning for cancer prognosis prediction

Morphological attributes from histopathological images and molecular profiles from genomic data are important information to drive diagnosis, prognosis, and therapy of cancers. By integrating these heterogeneous but complementary data, many multi-modal methods are proposed to study the complex mechanisms of cancers, and most of them achieve comparable or better results from previous single-modal methods. However, these multi-modal methods are restricted to a single task (e.g., survival analysis or grade classification), and thus neglect the correlation between different tasks. In this study, we present a multi-modal fusion framework based on multi-task correlation learning (MultiCoFusion) for survival analysis and cancer grade classification, which combines the power of multiple modalities and multiple tasks. Specifically, a pre-trained ResNet-152 and a sparse graph convolutional network (SGCN) are used to learn the representations of histopathological images and mRNA expression data respectively. Then these representations are fused by a fully connected neural network (FCNN), which is also a multi-task shared network. Finally, the results of survival analysis and cancer grade classification output simultaneously. The framework is trained by an alternate scheme. We systematically evaluate our framework using glioma datasets from The Cancer Genome Atlas (TCGA). Results demonstrate that MultiCoFusion learns better representations than traditional feature extraction methods. With the help of multi-task alternating learning, even simple multi-modal concatenation can achieve better performance than other deep learning and traditional methods. Multi-task learning can improve the performance of multiple tasks not just one of them, and it is effective in both single-modal and multi-modal data.

Self supervised contrastive learning for digital histopathology

Unsupervised learning has been a long-standing goal of machine learning and is especially important for medical image analysis, where the learning can compensate for the scarcity of labeled datasets. A promising subclass of unsupervised learning is self-supervised learning, which aims to learn salient features using the raw input as the learning signal. In this work, we tackle the issue of learning domain-specific features without any supervision to improve multiple task performances that are of interest to the digital histopathology community. We apply a contrastive self-supervised learning method to digital histopathology by collecting and pretraining on 57 histopathology datasets without any labels. We find that combining multiple multi-organ datasets with different types of staining and resolution properties improves the quality of the learned features. Furthermore, we find using more images for pretraining leads to a better performance in multiple downstream tasks, albeit there are diminishing returns as more unlabeled images are incorporated into the pretraining. Linear classifiers trained on top of the learned features show that networks pretrained on digital histopathology datasets perform better than ImageNet pretrained networks, boosting task performances by more than 28% in F1 scores on average. Interestingly, we did not observe a consistent correlation between the pretraining dataset site or the organ versus the downstream task (e.g., pretraining with only breast images does not necessarily lead to a superior downstream task performance for breast-related tasks). These findings may also be useful when applying newer contrastive techniques to histopathology data. Pretrained PyTorch models are made publicly available at https://github.com/ozanciga/self-supervised-histopathology.

Multi-Task Recurrent Modular Networks

We consider the models of deep multi-task learning with recurrent architectures that exploit regularities across tasks to improve the performance of multiple sequence processing tasks jointly. Most existing architectures are painstakingly customized to learn task relationships for different problems, which is not flexible enough to model the dynamic task relationships and lacks generalization abilities to novel test-time scenarios. We propose multi-task recurrent modular networks (MT-RMN) that can be incorporated in any multi-task recurrent models to address the above drawbacks. MT-RMN consists of a shared encoder and multiple task-specific decoders, and recurrently operates over time. For better flexibility, it modularizes the encoder into multiple layers of sub-networks and dynamically controls the connection between these sub-networks and the decoders at different time steps, which provides the recurrent networks with varying degrees of parameter sharing for tasks with dynamic relatedness. For the generalization ability, MT-RMN aims to discover a set of generalizable sub-networks in the encoder that are assembled in different ways for different tasks. The policy networks augmented with the differentiable routers are utilized to make the binary connection decisions between the sub-networks. The experimental results on three multi-task sequence processing datasets consistently demonstrate the effectiveness of MT-RMN.

Learning Centric Wireless Resource Allocation for Edge Computing: Algorithm and Experiment

Edge intelligence is an emerging network architecture that integrates sensing, communication, computing components, and supports various machine learning applications, where a fundamental communication question is: how to allocate the limited wireless resources (such as time, energy) to the simultaneous model training of heterogeneous learning tasks? Existing methods ignore two important facts: 1) different models have heterogeneous demands on training data; 2) there is a mismatch between the simulated environment and the real-world environment. As a result, they could lead to low learning performance in practice. This paper proposes the learning centric wireless resource allocation (LCWRA) scheme that maximizes the worst learning performance of multiple tasks. Analysis shows that the optimal transmission time has an inverse power relationship with respect to the generalization error. Finally, both simulation and experimental results are provided to verify the performance of the proposed LCWRA scheme and its robustness in real implementation.

Phrase embedding learning from internal and external information based on autoencoder

Phrase embedding can improve the performance of multiple NLP tasks. Most of the previous phrase-embedding methods that only use the external or internal semantic information of phrases to learn phrase embedding are challenging to solve the problem of data sparseness and have poor semantic presentation ability. To solve the above issues, in this paper, we propose an autoencoder-based method to combine pre-trained phrase embeddings and phrase component word embeddings into new phrase embeddings through complex non-linear transformations. This method uses both internal and external semantic information of phrases to generate new phrases with better semantic expression capabilities. This method can also generate well-represented phrase embeddings when only pre-trained component word embeddings are used as input to solve the problem of data sparseness effectively. We have designed two models for this method. The first one uses an FCNN(Fully Connected Neural Network) as the encoder and decoder, which we call AE-F. The second one uses the attention mechanism shared by the parameters of encoder and decoder to proportionally allocate the outputs of an LSTM and an FCNN, which we call it AE-ALF.We evaluated them in terms of phrase similarity and phrase classification and used two English datasets and two Chinese datasets. Experimental results show that AE-F and AE-ALF methods using pre-trained phrase embeddings and component word embeddings exceed 17 baseline methods, and AE-F and AE-ALF perform similarly. With only pre-trained component word embeddings, AE-F and AE-ALF also exceed most baseline methods, and AE-ALF performs better than AE-F.

Multi-task gradient descent for multi-task learning

Multi-Task Learning (MTL) aims to simultaneously solve a group of related learning tasks by leveraging the salutary knowledge memes contained in the multiple tasks to improve the generalization performance. Many prevalent approaches focus on designing a sophisticated cost function, which integrates all the learning tasks and explores the task-task relationship in a predefined manner. Different from previous approaches, in this paper, we propose a novel Multi-task Gradient Descent (MGD) framework, which improves the generalization performance of multiple tasks through knowledge transfer. The uniqueness of MGD lies in assuming individual task-specific learning objectives at the start, but with the cost functions implicitly changing during the course of parameter optimization based on task-task relationships. Specifically, MGD optimizes the individual cost function of each task using a reformative gradient descent iteration, where relations to other tasks are facilitated through effectively transferring parameter values (serving as the computational representations of memes) from other tasks. Theoretical analysis shows that the proposed framework is convergent under any appropriate transfer mechanism. Compared with existing MTL approaches, MGD provides a novel easy-to-implement framework for MTL, which can mitigate negative transfer in the learning procedure by asymmetric transfer. The proposed MGD has been compared with both classical and state-of-the-art approaches on multiple MTL datasets. The competitive experimental results validate the effectiveness of the proposed algorithm.

A Label-Enhanced Text Classification Model

In the field of text classification, most of the previous work only uses one-hot labels, ignoring the correlations between labels. The paper proposes a novel label-enhanced text classification model, which utilizes the semantic correlation between sentences and category labels in the Natural Language Processing (NLP) classification task to integrate label information. We measure the similarity between instance and the category with the correlations among the labels. We test the proposed model on text classification tasks in two levels: text classification (document level) and sentiment analysis (sentence level). Experimental results show that the label-enhanced text classification model achieves great performance in multiple text classification tasks. In addition, experiment results on the unbalanced datasets show that our model is able to mitigate the impact of unbalanced data in classification tasks.

A Hierarchical Discriminative Sparse Representation Classifier for EEG Signal Detection.

Classification of electroencephalogram (EEG) signal data plays a vital role in epilepsy detection. Recently sparse representation-based classification (SRC) methods have achieved the good performance in EEG signal automatic detection, by which the EEG signals are sparsely represented using a few active coefficients in the dictionary and classified according to the reconstruction criteria. However, most of SRC learn a linear dictionary for encoding, and cannot extract enough information and nonlinear relationship of data for classification. To solve this problem, a hierarchical discriminative sparse representation classification model (called HD-SRC) for EEG signal detection is proposed. Based on the framework of neural network, HD-SRC learns the hierarchical nonlinear transformation and maps the signal data into the nonlinear transformed space. Through incorporating this idea into label consistent K singular value decomposition (LC-KSVD) at the top layer of neural network, HD-SRC seeks discriminative representation together with dictionary, while minimizing errors of classification, reconstruction and discriminative sparse-code for pattern classification. By learning the hierarchical feature mapping and discriminative dictionary simultaneously, more discriminative information of data can be exploited. In the experiment the proposed model is evaluated on the Bonn EEG database, and the results show it obtains satisfactory classification performance in multiple EEG signal detection tasks.

Task partitioning in ants lacking discrete morphological worker subcastes

Task partitioning allows for efficient coordination of behavior in social insect colonies. Many task allocation studies focus on social insect species with discrete morphological worker subcastes, such as those possessing major and minor workers with strongly differentiated body plans. Much less is known about task partitioning among size-variable workers lacking discrete morphological subcastes. We investigated task fidelity and its correlation with worker size in Formica species with differing degrees of body size variation. During a mark-recapture study that focused on three worker tasks (honeydew collection, nest building, and protein foraging) across 2 days, 98.6% of 3570 recaptured workers exhibited task fidelity. In species with high levels of worker size variation, worker size is strongly correlated with task performance. This size-task correlation is weaker, but still present, in species with less variably sized workers. Our results suggest that Formica use size-based task partitioning, a form of morphological polyethism. We expect social insects with and without discrete morphological worker subcastes to differ in ontogeny, evolutionary history, and degree of behavioral flexibility. Identifying the scope of variation in task partitioning mechanisms will facilitate comparative studies, thereby elucidating evolutionary histories and outcomes of alternative strategies. Division of labor is thought to increase the efficiency and success of animal societies by allowing the performance of multiple tasks in parallel, rather than sequentially. We investigated the task partitioning strategy employed by an ecologically dominant ant genus characterized by continuous worker size variation but lacking discrete morphological worker subcastes. We show that several species employ a form of morphological polyethism, wherein individuals specialize in tasks according to body size. Size-mediated morphological polyethism in the absence of discrete morphological worker subcastes has been demonstrated in a few other social insects. However, this task allocation strategy is likely more common than previously recognized, as most insect societies display natural worker size variation. Further, we argue that it is important to distinguish between size- and subcaste-based morphological polyethism. Alternative task partitioning strategies are likely to impact animal societies’ resilience to environmental perturbations, which are becoming increasingly common in the face of global change.

Functional network connectivity (FNC)-based generative adversarial network (GAN) and its applications in classification of mental disorders

As a popular deep learning method, generative adversarial networks (GAN) have achieved outstanding performance in multiple classifications and segmentation tasks. However, the application of GANs to fMRI data is relatively rare. In this work, we proposed a functional network connectivity (FNC) based GAN for classifying psychotic disorders from healthy controls (HCs), in which FNC matrices were calculated by correlation of time courses derived from non-artefactual fMRI independent components (ICs). The proposed GAN model consisted of one discriminator (real FNCs) and one generator (fake FNCs), each has four fully-connected layers. The generator was trained to match the discriminator in the intermediate layers while simultaneously a new objective loss was determined for the generator to improve the whole classification performance. In a case for classifying 269 major depressive disorder (MDD) patients from 286 HCs, an average accuracy of 70.1% was achieved in 10-fold cross-validation, with at least 6% higher compared to the other 6 popular classification approaches (54.5–64.2%). In another application to discriminating 558 schizophrenia patients from 542 HCs from 7 sites, the proposed GAN model achieved 80.7% accuracy in leave-one-site-out prediction, outperforming support vector machine (SVM) and deep neural net (DNN) by 3%–6%. More importantly, we are able to identify the most contributing FNC nodes and edges with the strategy of leave-one-FNC-out recursively. To the best of our knowledge, this is the first attempt to apply the GAN model on the FNC-based classification of mental disorders. Such a framework promises wide utility and great potential in neuroimaging biomarker identification.

Scale-Wise Convolution for Image Restoration

While scale-invariant modeling has substantially boosted the performance of visual recognition tasks, it remains largely under-explored in deep networks based image restoration. Naively applying those scale-invariant techniques (e.g., multi-scale testing, random-scale data augmentation) to image restoration tasks usually leads to inferior performance. In this paper, we show that properly modeling scale-invariance into neural networks can bring significant benefits to image restoration performance. Inspired from spatial-wise convolution for shift-invariance, “scale-wise convolution” is proposed to convolve across multiple scales for scale-invariance. In our scale-wise convolutional network (SCN), we first map the input image to the feature space and then build a feature pyramid representation via bi-linear down-scaling progressively. The feature pyramid is then passed to a residual network with scale-wise convolutions. The proposed scale-wise convolution learns to dynamically activate and aggregate features from different input scales in each residual building block, in order to exploit contextual information on multiple scales. In experiments, we compare the restoration accuracy and parameter efficiency among our model and many different variants of multi-scale neural networks. The proposed network with scale-wise convolution achieves superior performance in multiple image restoration tasks including image super-resolution, image denoising and image compression artifacts removal. Code and models are available at: https://github.com/ychfan/scn_sr.

Multiobjective Multitasking Optimization Based on Incremental Learning

Multiobjective multitasking optimization (MTO) is an emerging research topic in the field of evolutionary computation. In contrast to multiobjective optimization, MTO solves multiple optimization tasks simultaneously. MTO aims to improve the overall performance of multiple tasks through knowledge transfer among tasks. Recently, MTO has attracted the attention of many researchers, and several algorithms have been proposed in the literature. However, one of the crucial issues, finding useful knowledge, has been rarely studied. Keeping this in mind, this article proposes an MTO algorithm based on incremental learning (EMTIL). Specifically, the transferred solutions (the form of knowledge) will be selected by incremental classifiers, which are capable of finding valuable solutions for knowledge transfer. The training data are generated by the knowledge transfer at each generation. Furthermore, the search space of the tasks will be explored by the proposed mapping (among tasks) approach, which helps these tasks to escape from their local Pareto Fronts. Empirical studies have been conducted on 15 MTO problems to assess the effectiveness of EMTIL. The experimental results demonstrate that EMTIL works more effectively for MTO compared to the existing algorithms.

NEWLSTM: An Optimized Long Short-Term Memory Language Model for Sequence Prediction

The long short-term memory (LSTM) model trained on the universal language modeling task overcomes the bottleneck of vanishing gradients in the traditional recurrent neural network (RNN) and shows excellent performance in processing multiple tasks generated by natural language processing. Although LSTM effectively alleviates the vanishing gradient problem in the RNN, the information will be greatly lost in the long distance transmission, and there are still some limitations in its practical use. In this paper, we propose a new model called NEWLSTM, which improves the LSTM model, and alleviates the defects of too many parameters in LSTM and the vanishing gradient. The NEWLSTM model directly correlates the cell state information with current information. The traditional LSTM’s input gate and forget gate are integrated, some components are deleted, the problems of too many LSTM parameters and complicated calculations are solved, and the iteration time is effectively reduced. In this paper, a neural network model is used to identify the relationship between input information sequences to predict the language sequence. The experimental results show that the improved new model is simpler than traditional LSTM models and LSTM variants on multiple test sets. NEWLSTM has better overall stability and can better solve the sparse words problem.

Scaling Camouflage: Content Disguising Attack Against Computer Vision Applications

Recently, deep neural networks have achieved state-of-the-art performance in multiple computer vision tasks, and become core parts of computer vision applications. In most of their implementations, a standard input preprocessing component called image scaling is embedded, in order to resize the original data to match the input size of pre-trained neural networks. This article demonstrates content disguising attacks by exploiting the image scaling procedure, which cause machine’s extracted content to be dramatically dissimilar with that before scaled. Different from previous adversarial attacks, our attacks happen in the data preprocessing stage, and hence they are not subject to specific machine learning models. To achieve a better deceiving and disguising effect, we propose and implement three feasible attack approaches with $L_0$ L 0 -, $L_2$ L 2 - and $L_{\infty }$ L ∞ -norm distance metrics. We have conducted a comprehensive evaluation on various image classification applications, including three local demos and two remote proprietary services. We also investigate the attack effects on a YOLO-v3 object detection demo. Our experimental results demonstrate successful content disguising against all of them, which validate our approaches are practical.

Supporting Multitracking Performance With Novel Visual, Auditory, and Tactile Displays

This article investigates performance in multiple concurrent tracking tasks with multisensory displays in a driving context. In many work domains, such as driving, aviation, process control, and medicine, humans perform “tracking” tasks that involve observing continuous variables and providing a control input to achieve and maintain satisfactory levels in those variables. Performance in multiple concurrent tracking tasks (“multitracking”) was studied in driving-like scenarios that challenged participants to track established targets for lateral (lane position) and longitudinal (speed) variables. Novel speed displays were developed to engage nontraditional sensory modalities (e.g., ambient-visual, auditory, or tactile) with relative speed conveyed through simple or multidimensional signal encoding methods. Participants’ speed-tracking and lane-tracking performances were measured concurrently and compared across display configurations within-subjects. Results showed lane tracking performance to be unaffected by display configuration; however, speed tracking and overall performance were significantly improved with novel displays, compared to the baseline configuration. Redundantly encoded auditory displays best-supported multitracking performance, but redundantly encoded tactile displays were not as beneficial as were simple encodings. These results provide insight into the human information processing of semicontinuous multisensory displays and can inform display design in driving and other visually demanding work contexts.

A Preliminary Study of Dual-Task Training Using Virtual Reality: Influence on Walking and Balance in Chronic Poststroke Survivors

BackgroundStroke is a leading cause of death and disability in the Western world, and leads to impaired balance and mobility. Objective: To investigate the feasibility of using a Virtual Reality-based dual task of an upper extremity while treadmill walking, to improve gait and functional balance performance of chronic poststroke survivors. MethodsTwenty-two individuals chronic poststroke participated in the study, and were divided into 2 groups (each group performing an 8-session exercise program): 11 participated in dual-task walking (DTW), and the other 11 participated in single-task treadmill walking (TMW). The study was a randomized controlled trial, with assessors blinded to the participants’ allocated group. Measurements were conducted at pretest, post-test, and follow-up. Outcome measures included: the 10-m walking test (10 mW), Timed Up and Go (TUG), the Functional Reach Test (FRT), the Lateral Reach Test Left/Right (LRT-L/R); the Activities-specific Balance Confidence (ABC) scale, and the Berg Balance Scale(BBS). ResultsImprovements were observed in balance variables: BBS, FRT, LRT-L/R, (P < .01) favoring the DTW group; in gait variables: 10 mW time, also favoring the DTW group (P < .05); and the ABC scale (P < .01). No changes for interaction were observed in the TUG. ConclusionsThe results of this study demonstrate the potential of VR-based DTW to improve walking and balance in people after stroke; thus, it is suggested to combine training sessions that require the performance of multiple tasks at the same time.

Feature extraction of four-class motor imagery EEG signals based on functional brain network

Objective. A motor-imagery-based brain–computer interface (MI-BCI) provides an alternative way for people to interface with the outside world. However, the classification accuracy of MI signals remains challenging, especially with an increased number of classes and the presence of high variations with data from multiple individual people. This work investigates electroencephalogram (EEG) signal processing techniques, aiming to enhance the classification performance of multiple MI tasks in terms of tackling the challenges caused by the vast variety of subjects. Approach. This work introduces a novel method to extract discriminative features by combining the features of functional brain networks with two other feature extraction algorithms: common spatial pattern (CSP) and local characteristic-scale decomposition (LCD). After functional brain networks are established from the MI EEG signals of the subjects, the measures of degree in the binary networks are extracted as additional features and fused with features in the frequency and spatial domains extracted by the CSP and LCD algorithms. A real-time BCI robot control system is designed and implemented with the proposed method. Subjects can control the movement of the robot through four classes of MI tasks. Both the BCI competition IV dataset 2a and real-time data acquired in our designed system are used to validate the performance of the proposed method. Main results. As for the offline data experiment results, the average classification accuracy of the proposed method reaches 79.7%, outperforming the majority of popular algorithms. Experimental results with real-time data also prove the proposed method to be highly promising in its real-time performance. Significance. The experimental results show that our proposed method is robust in extracting discriminative brain activity features when performing different MI tasks, hence improving the classification accuracy in four-class MI tasks. The high classification accuracy and low computational demand show a considerable practicality for real-time rehabilitation systems.

Multi-task prioritization during the performance of a postural-manual and communication task.

Individuals are often required to complete two tasks simultaneously, such as walking while talking. Although the influence of performing a cognitive task during upright standing has been studied, less is known regarding how individuals prioritize the performance of multiple goal-directed tasks. The purpose of this study was to examine how young adults prioritize the performance of two goal-directed tasks while standing: generating speech (communicating) while performing a postural-manual task. Fifteen young adult female participants completed four conditions of the postural-manual task, quietly and while talking; a control condition of speech while standing was also completed. Task accuracy was defined as holding a block within an opening without hitting the perimeter and speaking clearly. Participants generally prioritized accurate communication over accurate completion of the postural-manual task, as evidenced by modifying their manual and postural strategies and decreasing manual accuracy.