Related Tasks Research Articles

Attentional warnings caused by driver monitoring systems: How often do they appear and how well are they understood?

The present study investigated the effects of a driver monitoring system that triggers attention warnings in case distraction is detected. Based on the EuroNCAP protocol, distraction could either be long glances away from the forward roadway (≥3s) or visual attention time sharing (>10 cumulative seconds within a 30 s time interval). In a series of manual driving simulator drives, 30 participants completed both driving related tasks (e.g., changing multiple lanes in dense traffic) and non-driving related tasks (e.g., infotainment operations). Results of warning frequencies revealed that visual attention time sharing warnings occurred more frequently than long distraction warnings. Moreover, there was a large number of attention warnings during driving related tasks. Results also revealed that participants’ mental models tended to be less accurate when it came to understanding of the visual attention time sharing warnings as compared to the long distraction warnings, which were understood more accurately. Based on these observations, the work discusses the applicability and design of driver monitoring warnings.

Multi-word expressions for Russian L2 learners: corpora-based selection with expert verification

The article describes the experience of creating a corpus-based list of the most relevant multi-word expressions for Russian L2 learners, distributed across the levels of the Common European Framework of Reference for Languages (CEFR) from A1 to C1. Modern linguistic and cognitive research shows that our speech is patterned and largely consists of stable segments. This fact is supported by the linguodidactic idea of teaching not isolated language units but their combinations of different nature. However, the selection and ranking of multi-word expressions based on language proficiency levels is constrained by the difficulty of automatically extracting them from a corpus of texts and estimating their frequency, as well as disagreements in defining the boundaries, linguistic nature, and terminology of multi-word expressions. This article describes the experience of compiling a list of the most valuable fixed-type multi-word expressions from various sources: two types of existing CEFR-graded vocabulary lists for Russian L2 learners – lexical minimums for the TORFL (Test of Russian as a Foreign Language) system and Russian KELLY (KEywords for Language Learning for Young and adults alike); the most frequent n-grams from the RuFoLa – Russian L2 textbook corpus and from the Russian Web corpus of internet texts; list of discourse formulas from the «Pragmaticon» project. The CEFR level of each multi-word expression is predicted using the frequency-based Max Delta measure, and its effectiveness is subsequently validated through annotation by multiple experts. The resulting list of multi-word expressions contains 1645 entries from A1 to C1 levels. The proposed version of the list has been implemented into an automated text analysis system for learners of Russian as a Foreign Language and can be useful for a wide range of professionals in the preparation of educational content for foreign language learners. The suggested Max Delta measure has demonstrated a high degree of agreement with expert evaluations within proficiency levels A1-B1. This signifies the importance of further exploring its potential in addressing related practical tasks and in selecting language learning content derived for other languages.

Point cloud registration via sampling-based evolutionary multitasking

Point cloud registration (PCR) is an important task in both computer vision and pattern recognition. However, since the success of most existing two-stage PCR methods largely depends on the implicit assumption that the first coarse registration stage can provide an approximately optimal initial transformation for the second fine stage, these methods suffer from low success rate when the assumption is not satisfied. Evolutionary multitasking can find better solutions for multiple tasks simultaneously by transferring knowledge among multiple related tasks. Inspired by evolutionary multitasking, this study proposes a sampling-based multitask optimization method for PCR, which solves the problem of low registration success rate by sharing knowledge between two related tasks generated by a dataset. Specifically, one task is for roughly aligning the sampled point cloud, and the other is for the original unsampled point cloud. The main purpose of the first task is to continuously provide useful knowledge to the second task so that it can achieve precise alignment. Then, a new knowledge transfer mechanism guided by historical information is proposed, which can effectively adjust the search behavior of the original unsampled task. In addition, we develop a dynamic task complement strategy to complement the accuracy that may be lost by downsampling. The results on five different datasets show that the proposed method can achieve higher registration accuracy with higher success rate compared to other state-of-the-art PCR methods.

A demonstration trajectory segmentation approach for wheelchair‐mounted assistive robots

AbstractSegmentation of demonstration trajectories and learning the contained motion primitives can effectively enhance the assistive robot's intelligence to flexibly reproduce learnt tasks in an unstructured environment. With the aim to conveniently and accurately segment demonstration trajectories, a novel demonstration trajectory segmentation approach is proposed based on the beta process autoregressive hidden Markov model (BP‐AR‐HMM) algorithm and generalised time warping (GTW) algorithm aiming to enhance the segmentation accuracy utilising acquired demonstration data. This approach first adopts the GTW algorithm to align the multiple demonstration trajectories for the same task. Then, it adopts the BP‐AR‐HMM algorithm to segment the demonstration trajectories, acquire the contained motion primitives, and establish the related task library. This segmentation approach is validated on the 6‐degree‐of‐freedom JACO robotic arm by assisting users to accomplish a holding water glass task and an eating task. The experimental results show that the motion primitives within the trajectories can be correctly segmented with a high segmentation accuracy.

Relationship of Smartphone Use and Job Performance: Insights from Electronic Media Industry in Pakistan

This research article explores the relationship between smartphone usage and jobperformance among professionals working in the leading electronic media channels inPakistan. Furthermore, it also examines how gender, education, and age act as moderators in the relationship between smartphone usage and job performance. For this purpose, asurvey was carried out involving professionals belonging to the electronic media industryin Pakistan. The findings revealed a positive and significant relationship between their smartphone usage and job performance. The results also showed that the gender of employees did not moderate this relationship, while the education and the age were foundto be significant moderators. The study will help the media owners in devising thepolicies and strategies to facilitate their workers in easy utilization of smartphones for work related tasks.

Analisis sistem pengendalian internal prosedur pengeluaran kas: COSO ICIF 2013

Financial management plays a crucial role in the survival and growth of companies. Cash disbursement is a common transaction in companies, necessitating effective internal control systems to prevent fraud or misappropriation. This study aims to analyze the internal control system and evaluate the internal control system over cash disbursements at PT WE, a mining consulting company in Yogyakarta. The research method used is descriptive-evaluative with a qualitative approach. Data were collected through direct observation and interviews with relevant staff. Evaluation was conducted based on the 17 principles within the five main components of the COSO Internal Control—Integrated Framework 2013. The results indicate that PT WE has implemented some internal control principles effectively, but there are still weaknesses in interdepartmental coordination, risk identification, and cash disbursement procedures. Based on the findings, it is recommended that PT WE improve coordination and communication between departments, systematically identify risks, and enhance cash disbursement procedures by establishing clear protocols, segregating related tasks, and periodically evaluating the internal control system.

A Review on Alzheimer's Disease Classification Using Deep Learning

In recent years, there has been a substantial amount of research dedicated to using Deep Learning (DL) methods for the classification of Alzheimer's disease (AD) and other related tasks, specifically focusing on magnetic resonance imaging (MRI) data. According to a comprehensive analysis of recent studies, it seems that deep learning models, especially those that include the creation of different structures, have significant potential to improve the precision of identifying and classifying Alzheimer's disease at an early stage. This work aims to emphasize the importance of effective data preparation tactics and feature learning approaches, as well as the investigation of hybrid models using diverse deep learning technologies. This study primarily focuses on doing performance analysis of deep learning algorithms using the latest approaches. Finally, provide a concise overview and analysis of several methods that might enhance the effectiveness of identification and classification using deep learning.

MOREM: An evolutionary multitasking optimization algorithm for multi-objective recommendations

Multi-objective evolutionary algorithms (MOEAs) have been successfully applied in recommender systems, where the recommendation is modeled as a multi-objective optimization problem by considering both accuracy and non-accuracy metrics. However, to recommend to all users, the population individuals in most of the exiting MOEA-based recommendation algorithms are usually encoded in the form of matrix encoding, and with the increased number of users and items, it will lead to a large search space. To this end, in this paper, we propose a novel evolutionary multitasking algorithm MOREM to tackle the challenge of multi-objective recommendations, where a related auxiliary task is constructed by both user and item reduction to help solving the complex original task by knowledge transfer. Specifically, a task generation strategy for creating auxiliary task is firstly proposed to cluster users by user-rating information (i.e., user reduction), where only a part of important items (i.e., item reduction) are recommended for the central user within each cluster with the aim to greatly reduce the size of the matrix encoding. In addition, a novel knowledge transfer mechanism is proposed, which can effectively achieve knowledge migration between the two tasks. Experimental results on real-world datasets show that MOREM outperforms several state-of-the-art algorithms for multi-objective recommendations.

Autism spectrum disorders detection based on multi-task transformer neural network

Autism Spectrum Disorders (ASD) are neurodevelopmental disorders that cause people difficulties in social interaction and communication. Identifying ASD patients based on resting-state functional magnetic resonance imaging (rs-fMRI) data is a promising diagnostic tool, but challenging due to the complex and unclear etiology of autism. And it is difficult to effectively identify ASD patients with a single data source (single task). Therefore, to address this challenge, we propose a novel multi-task learning framework for ASD identification based on rs-fMRI data, which can leverage useful information from multiple related tasks to improve the generalization performance of the model. Meanwhile, we adopt an attention mechanism to extract ASD-related features from each rs-fMRI dataset, which can enhance the feature representation and interpretability of the model. The results show that our method outperforms state-of-the-art methods in terms of accuracy, sensitivity and specificity. This work provides a new perspective and solution for ASD identification based on rs-fMRI data using multi-task learning. It also demonstrates the potential and value of machine learning for advancing neuroscience research and clinical practice.

A Novel Online Hydrological Data Quality Control Approach Based on Adaptive Differential Evolution

The quality of hydrological data has a significant impact on hydrological models, where stable and anomaly-free hydrological time series typically yield more valuable patterns. In this paper, we conduct data analysis and propose an online hydrological data quality control method based on an adaptive differential evolution algorithm according to the characteristics of hydrological data. Taking into account the characteristics of continuity, periodicity, and seasonality, we develop a Periodic Temporal Long Short-Term Memory (PT-LSTM) predictive control model. Building upon the real-time nature of the data, we apply the Adaptive Differential Evolution algorithm to optimize PT-LSTM, creating an Online Composite Predictive Control Model (OCPT-LSTM) that provides confidence intervals and recommended values for control and replacement. The experimental results demonstrate that the proposed data quality control method effectively manages data quality; detects data anomalies; provides suggested values; reduces reliance on manual intervention; provides a solid data foundation for hydrological data analysis work; and helps hydrological personnel in water resource scheduling, flood control, and other related tasks. Meanwhile, the proposed method can also be applied to the analysis of time series data in other industries.

Identify and mitigate bias in electronic phenotyping: A comprehensive study from computational perspective

Electronic phenotyping is a fundamental task that identifies the special group of patients, which plays an important role in precision medicine in the era of digital health. Phenotyping provides real-world evidence for other related biomedical research and clinical tasks, e.g., disease diagnosis, drug development, and clinical trials, etc. With the development of electronic health records, the performance of electronic phenotyping has been significantly boosted by advanced machine learning techniques. In the healthcare domain, precision and fairness are both essential aspects that should be taken into consideration. However, most related efforts are put into designing phenotyping models with higher accuracy. Few attention is put on the fairness perspective of phenotyping. The neglection of bias in phenotyping leads to subgroups of patients being underrepresented which will further affect the following healthcare activities such as patient recruitment in clinical trials. In this work, we are motivated to bridge this gap through a comprehensive experimental study to identify the bias existing in electronic phenotyping models and evaluate the widely-used debiasing methods’ performance on these models. We choose pneumonia and sepsis as our phenotyping target diseases. We benchmark 9 kinds of electronic phenotyping methods spanning from rule-based to data-driven methods. Meanwhile, we evaluate the performance of the 5 bias mitigation strategies covering pre-processing, in-processing, and post-processing. Through the extensive experiments, we summarize several insightful findings from the bias identified in the phenotyping and key points of the bias mitigation strategies in phenotyping.

How do visual and cognitive non-driving related tasks affect drivers’ visual attention and takeover performance in conditionally automated driving?

This study examines changes in visual attention during drivers’ engagement in visual or cognitive NDRTs during conditionally automated driving, and determines how these changes affect takeover performance. Seventy-five participants took part in a driving simulator study, performing three pre-takeover tasks: the three pre-takeover tasks: an auditory-imagery-verbal task (cognitive NDRT); a video-watching task (visual NDRT); and a monitoring task (baseline/non NDRT). Also, there were two hazardous events (breakdown or sudden merging of the vehicle ahead) leading to takeover requests issued with 7-s or 5-s lead times. The results revealed that NDRTs negatively affected visual attention, which caused lower saccade frequency between different areas of interest and shorter saccade amplitude. Drivers paid more visual attention to the road ahead and in-vehicle information system when performing cognitive and visual NDRTs, respectively. The visual attention of drivers performing NDRTs negatively affected takeover performance (e,g longer reaction time, heavier maximal brake pedal input, etc.). The reduction in takeover request lead time impaired takeover performance. The findings will support the design of eye tracker-based “out-of-loop” discrimination techniques and human-machine interaction interfaces in automated vehicles. This study contributes to the literature by examining how different types of NDRTs affect takeover performance, specifically focusing on drivers’ visual attention.

Interpretable multi-task neural network modeling and particle swarm optimization of process parameters in laser welding

The process parameters directly affect the quality and performance of laser welding (LW) by shaping the molten pool appearance. However, with the increase in the dimensions of process parameters and the number of samples, the relationship between process parameters and molten pool often exhibits high levels of uncertainty and nonlinearity. Machine learning (ML) techniques have been used for complex LW systems modeling, but they may be limited and unable to fully capture the complex relationships when faced with complex systems. Additionally, traditional ML techniques may lack interpretability within the model, making it difficult to understand the input features and their impact on the output. To address these issues, a method that combines an interpretable multi-task neural network (MTNN) and particle swarm optimization (PSO) is proposed. Firstly, an LW experimental platform is constructed, and the orthogonal design is conducted for data collection. In the data preparation stage, the curriculum learning strategy is employed to reorganize the raw experimental data. Subsequently, the MTNN is constructed to automatically learn feature representations among input process parameters, capture nonlinear relationships between data, and share weights across multiple related tasks to improve efficiency and performance. Furthermore, the PSO is designed to determine the constraints and objective functions. In addition, the shapley additive explanation (SHAP) method is utilized to calculate the importance of process parameters and enhance the interpretability. Finally, the optimization results are validated, and the results demonstrate that the optimized results of the proposed method have achieved satisfactory performance compared to the traditional ML methods.

Associative Learning of an Unnormalized Successor Representation.

The successor representation is known to relate to temporal associations learned in the temporal context model (Gershman etal., 2012), and subsequent work suggests a wide relevance of the successor representation across spatial, visual, and abstract relational tasks. I demonstrate that the successor representation and purely associative learning have an even deeper relationship than initially indicated: Hebbian temporal associations are an unnormalized form of the successor representation, such that the two converge on an identical representation whenever all states are equally frequent and can correlate highly in practice even when the state distribution is nonuniform.

Physiological and cognitive responses to load carriage in the cold

Load carriage is a vital role related task for military, emergency service, and search and rescue personnel, through the transport of critical equipment. From an external validity perspective, a common oversight in previous research is the effect of multiple stressors during load carriage; for example, the inclusion of adverse environmental conditions and/or concurrent cognitive tasks. Therefore, this study aimed to quantify physiological and cognitive changes over time during military specific fast-paced lower-weight load carriage in a cold environment. With approval from the University of Chichester’s Ethics Committee, nine participants completed a load carriage task consisting of 20 min walking at 5.1 km∙h-1, 40 min walking at 6.5 km∙h-1, followed by 8 × 9 s shuttles running at 11 km∙h-1 in an environmental temperature of −10°C. Participants wore overalls whilst carrying 16 kg of external load (10 kg belt-webbing, 6 kg body armour). Throughout, measurements were taken of heart rate, core and skin temperatures, expired gas (VO2), and ratings of physical exertion and thermal comfort. Six military-specific auditory n-back tasks (MSANT) were completed across the trial. The rating scale of mental effort (RSME) quantified mental exertion during the load carriage task. Statistical differences within variables were evaluated using a one-way repeated measures ANOVA with post-hoc analysis employed to identify differences between timepoints. Heart rate and VO2 behaved as expected, with increases after the 5.1 to 6.5 km∙h-1 speed escalation (Pbonf< 0.001); heart rate also increased during the shuttles compared to the first 6.5 km∙h-1 collection (Pbonf = 0.003). A main effect rise for core temperature was evident (Pbonf < 0.001), increasing after 25 min at 6.5 km∙h-1 (Pbonf = 0.009). Skin temperature presented a main effect (Pbonf < 0.001) with an initial decrease and plateau occurring after 15 min. There was no main effect for MSANT or RSME at MSANT completion timepoints (P’s= 0.935 and 0.368 respectively). However, during non-MSANT completion timepoints, RSME was greater from 25 min onwards compared to 5 min (Pbonf < 0.05) signifying an increase in inter-task cognitive demand, oppose to a period of relative mental recovery. This reduced efficiency could be exacerbated during concurrent or extended physical and cognitive tasks. This study is the first to examine physiological and cognitive responses to fast-paced lower-weight load carriage in the cold, progressing understanding of occupational performance in representative environments. These performance data can inform training protocols and operation approaches by highlighting higher inter-task cognitive demand, potentially emphasising an avenue for mitigation strategies.

Stitched vision transformer for age-related macular degeneration detection using retinal optical coherence tomography images.

Age-related macular degeneration (AMD) is an eye disease that leads to the deterioration of the central vision area of the eye and can gradually result in vision loss in elderly individuals. Early identification of this disease can significantly impact patient treatment outcomes. Furthermore, given the increasing elderly population globally, the importance of automated methods for rapidly monitoring at-risk individuals and accurately diagnosing AMD is growing daily. One standard method for diagnosing AMD is using optical coherence tomography (OCT) images as a non-invasive imaging technology. In recent years, numerous deep neural networks have been proposed for the classification of OCT images. Utilizing pre-trained neural networks can speed up model deployment in related tasks without compromising accuracy. However, most previous methods overlook the feasibility of leveraging pre-existing trained networks to search for an optimal architecture for AMD staging on a new target dataset. In this study, our objective was to achieve an optimal architecture in the efficiency-accuracy trade-off for classifying retinal OCT images. To this end, we employed pre-trained medical vision transformer (MedViT) models. MedViT combines convolutional and transformer neural networks, explicitly designed for medical image classification. Our approach involved pre-training two distinct MedViT models on a source dataset with labels identical to those in the target dataset. This pre-training was conducted in a supervised manner. Subsequently, we evaluated the performance of the pre-trained MedViT models for classifying retinal OCT images from the target Noor Eye Hospital (NEH) dataset into the normal, drusen, and choroidal neovascularization (CNV) classes in zero-shot settings and through five-fold cross-validation. Then, we proposed a stitching approach to search for an optimal model from two MedViT family models. The proposed stitching method is an efficient architecture search algorithm known as stitchable neural networks. Stitchable neural networks create a candidate model in search space for each pair of stitchable layers by inserting a linear layer between them. A pair of stitchable layers consists of layers, each selected from one input model. While stitchable neural networks had previously been tested on more extensive and general datasets, this study demonstrated that stitching networks could also be helpful in smaller medical datasets. The results of this approach indicate that when pre-trained models were available for OCT images from another dataset, it was possible to achieve a model in 100 epochs with an accuracy of over 94.9% in classifying images from the NEH dataset. The results of this study demonstrate the efficacy of stitchable neural networks as a fine-tuning method for OCT image classification. This approach not only leads to higher accuracy but also considers architecture optimization at a reasonable computational cost.

BioInstruct: instruction tuning of large language models for biomedical natural language processing.

To enhance the performance of large language models (LLMs) in biomedical natural language processing (BioNLP) by introducing a domain-specific instruction dataset and examining its impact when combined with multi-task learning principles. We created the BioInstruct, comprising 25005 instructions to instruction-tune LLMs (LLaMA 1 and 2, 7B and 13B version). The instructions were created by prompting the GPT-4 language model with 3-seed samples randomly drawn from an 80 human curated instructions. We employed Low-Rank Adaptation (LoRA) for parameter-efficient fine-tuning. We then evaluated these instruction-tuned LLMs on several BioNLP tasks, which can be grouped into 3 major categories: question answering (QA), information extraction (IE), and text generation (GEN). We also examined whether categories (eg, QA, IE, and generation) of instructions impact model performance. Comparing with LLMs without instruction-tuned, our instruction-tuned LLMs demonstrated marked performance gains: 17.3% in QA on average accuracy metric, 5.7% in IE on average F1 metric, and 96% in Generation tasks on average GPT-4 score metric. Our 7B-parameter instruction-tuned LLaMA 1 model was competitive or even surpassed other LLMs in the biomedical domain that were also fine-tuned from LLaMA 1 with vast domain-specific data or a variety of tasks. Our results also show that the performance gain is significantly higher when instruction fine-tuning is conducted with closely related tasks. Our findings align with the observations of multi-task learning, suggesting the synergies between 2 tasks. The BioInstruct dataset serves as a valuable resource and instruction tuned LLMs lead to the best performing BioNLP applications.

A sequential meta-transfer (SMT) learning to combat complexities of physics-informed neural networks: Application to composites autoclave processing

Physics-Informed Neural Networks (PINNs) have gained popularity in solving nonlinear partial differential equations (PDEs) via integrating physical laws into the training of neural networks, making them superior in many scientific and engineering applications. However, conventional PINNs still fall short in accurately approximating the solution of complex systems with strong nonlinearity, especially in long temporal domains. Besides, since PINNs are designed to approximate a specific realization of a given PDE system, they lack the necessary generalizability to efficiently adapt to new system configurations. This entails computationally expensive re-training from scratch for any new change in the system. To address these shortfalls, in this work a sequential meta-transfer (SMT) learning framework is proposed, offering a unified solution for both fast training and efficient adaptation of PINNs in highly nonlinear systems with long temporal domains. Specifically, the framework decomposes PDE's time domain into smaller time segments to create “easier” PDE problems for PINNs training. Then for each time interval, a meta-learner is assigned and trained to achieve an optimal initial state for rapid adaptation to a range of related tasks. Transfer learning principles are then leveraged across time intervals to further reduce the computational cost. Through a composites autoclave processing case study, it is shown that SMT is clearly able to enhance the adaptability of PINNs while significantly reducing computational cost, by a factor of 100.

Past materials, past minds: The philosophy of cognitive paleoanthropology

AbstractThe philosophy of cognitive paleoanthropology involves three related tasks: (1) asking what inferences might be drawn from the paleontological and archaeological records to past cognition, behavior and culture; (2) constructing synthetic accounts of the evolution of distinctive hominin capacities; (3) exploring how results from cognitive paleoanthropology might inform philosophy. We introduce some distinctive cognitive paleoanthropological inferences and discuss their epistemic standing, before considering how attention to the material records and the practice of paleoanthropology can inform and transform philosophical approaches.

Using reaction time procedures to assess implicit attitudes toward violence in a nonconvicted male sample.

In this study, we sought to capture implicit attitudes toward violence by administering response latency measures. We then examined their associations with explicit (e.g., assessed with self-report) attitudes toward violence and self-reported violent behavior in a combined sample of males from a Canadian university and males from the general community (N = 251; 156 students and 95 community members). To date, there have been mixed findings regarding these associations; some of this inconsistency may be due to the difficulty in accurately conceptualizing and assessing implicit attitudes toward violence. Therefore, we administered three response latency measures to assess this construct: a violence evaluation implicit association test (VE-IAT), a personalized VE-IAT (P-VE-IAT), and a violence evaluation relational responding task, along with three self-report measures of explicit attitudes toward violence and three self-report measures of violent behavior. More positive implicit attitudes toward violence were related to more positive explicit attitudes toward violence (for VE-IAT and P-VE-IAT; r = 0.18 to 0.22), greater likelihood of violence (for VE-IAT; r = 0.18 and for P-VE-IAT; r = 0.16), and greater propensity for violence (for the VE-IAT; r = 0.16). All measures of explicit attitudes toward violence and violent behavior were moderately to strongly associated with one another (r = 0.42 to 0.81). Furthermore, implicit attitudes toward violence explained additional variance in some violent outcomes above explicit attitudes alone. Our findings suggest that scores on certain reaction time measures are important for understanding likelihood and propensity for violence, especially when combined with explicit attitude measures.