Dual Learning Model Research Articles

Urban climate adaptability and green total-factor productivity: Evidence from double dual machine learning and differences-in-differences techniques

Climate change has increasingly become a significant challenge to sustainable socio-economic development, and climate adaptation is a key issue that relevant research focuses on regional sustainable development models. By employing panel data between 2007 and 2020 from 284 Chinese prefecture-level cities, this study adopts quasi-experimental methods, including a difference-in-differences design and double dual machine learning model, to study the impact of climate adaptability on green regional sustainable development. Empirical results confirm that the pilot policy of building climate-resilient cities significantly improves urban green total-factor productivity. Difference-in-difference models (derived from entropy-weight and propensity score matching) and double dual learning models also support the improving effect of regional green total-factor productivity after policy intervention. The digital economy has strengthened the green development effect of pilot policies for building climate-adaptive cities. In addition, policy interventions to build climate-adaptive cities promote green urban development by optimizing industrial development structures and enhancing economic growth resilience. In addition, climate adaptability can also attract highly skilled talent and high-quality enterprises, facilitate science and technological progress in urban areas, and thus promoting the green development of cities in China. This study objectively evaluates the effects of climate policies and provides insights for global adaptation to climate change and optimization of public policies.

Weaving game and task into content-language integration: Proposing a game-mediated dual learning model

Content-based instruction, task-based language teaching, and game-mediated learning are three pedagogical approaches that are perceived as effective in second and foreign language education. These approaches share common ground in a sociocultural and functional view of language learning and popularity as engaging classroom strategies in elementary, middle, and high school (K-12) contexts. However, challenges may exist in ensuring content and language dual learning goals, designing language learning tasks, or implementing game-mediated activities. Therefore, this study seeks to describe the conceptual and methodological alignment among these approaches by reviewing recent research. Moreover, a game-mediated dual learning model is proposed to pinpoint a series of design components for language educators to consider when they integrate the aforementioned three approaches.

Half Quadratic Dual Learning for Fuzzy Multiconcepts of Partially-Observed Images

For partially-observed data with large continuous occlusion or specific outlier noise, it is difficult to represent them by using L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> -, L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> -, and L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2,1</sub> -norms, not to mention learning multiconcept mapping from them. Furthermore, multiconcept learning (i.e., multilabel learning) and sample-correlation learning are inseparable as they provide guiding information to each other. Nonetheless, when most sample relations are unknown/biased, detecting multiconcepts associated with partially-observed data becomes even more difficult. In view of such challenges, this study presents a dual learning model that can simultaneously learn multiconcepts, sample relations, and data representation while partially-observed data and partially-observed sample relations are present. Moreover, half quadratic (HQ) loss functions along with half quadratic optimization and their gradient forms are developed in this study to model unobserved information. Different types of partially-observed sources are jointly hinged together to compensate each other during learning. This allows the system to estimate accurate multiconcepts, sample relations, and data representation. Finally, fuzzy scores for multiconcepts can be learnt and generated. Thus, a labelled data instance can utilize such fuzzy scores to rank its concepts. Experiments on open images datasets were carried out for evaluating the performance. Analytical results showed that the proposed model was capable of conquering multiconcept learning problems while data were partially observed, better than the baselines, thereby demonstrating the effectiveness of the proposed idea.

Learning Patterns, Personality Traits and Motivational Profile of Students of Dual Learning Modality

The aim of this research article is to identify the profile of students enrolled in dual learning model and standard learning model based on personality traits, learning patterns and motivational orientation. Participants included 212 freshmen enrolled dual learning model (N=97) and standard model (N=115) in the Primary Education Degree. The participants completed 3 questionnaires: (1) Learning Combination Inventory (Johnston &amp; Dainton, 1996), (2) Big-Five Personality Inventory (Bermúdez, 1995), and (3) Motivational Profile Inventory (Corral, Arribas &amp; Fernández, 2010). The results showed that the students who enrolled in dual learning modality tend to be less open, more emotionally stable and take care of their relationships to a greater extent. The two later characteristics are essential for the typology of the learning modality because it is more complex to work and learn in two different environments than one as standard learning. Despite these results, the groups of both learning modalities are similar and therefore possible future lines are considered. Future studies should be conducted with much larger samples and longitudinal design that allows corroborating (or not) that the students who study in dual learning modality are different or can vary their motivational orientations or improve their learning patterns during the course of their training.

A Dual Learning Model for Vehicle Trajectory Prediction

Automated vehicles and advanced driver-assistance systems require an accurate prediction of future traffic scene states. The tendency in recent years has been to use deep learning approaches for accurate trajectory prediction but these approaches suffer from computational complexity, dependency on a specific environment/dataset, and lack of insight into vehicle interactions. In this paper, we aim to address these limitations by proposing a Dual Learning Model (DLM) using lane occupancy and risk maps for vehicle trajectory prediction. To understand the spatial interactions of road users, make the model independent of the environment, and consider inter-vehicle distances, we embed an Occupancy Map (OM) into the trajectory prediction model. We also utilise a traffic scene Risk Map (RM) to explicitly consider a comprehensive definition of risk based on Time-to-Collision in the traffic scene. These two features employed in the encoder-decoder architecture improve system accuracy with less complexity and provide insight into the interaction between all road users. The experiment has been conducted on two different naturalistic highway driving datasets (i.e., NGSIM and HighD) demonstrating algorithm independence from a single environment. Comparison results indicate that the DLM achieves a more accurate trajectory prediction with a less complex structure compared with existing approaches in terms of RMS prediction error, which indicates the effectiveness of DLM in such a context.

JOVACET Volume 2 Issue 2 (2019)

JOVACET Volume 2 Issue 2 (2019)

Dual memory network model for sentiment analysis of review text

In sentiment analysis of product reviews, both user and product information are proven to be useful. Current works handle user profile and product information in a unified model which may not be able to learn salient features of users and products effectively. In this work, we propose a dual user and product memory network (DUPMN) model to learn user profiles and product information for reviews classification using separate memory networks. Then, the two representations are used jointly for sentiment analysis. The use of separate models aims to capture user profiles and product information more effectively. Comparing with state-of-the-art unified prediction models, evaluations on three benchmark datasets (IMDB, Yelp13, and Yelp14) show that our dual learning model gives performance gain of 0.6%, 1.2%, and 0.9%, respectively. The improvements are also deemed very significant measured by p-values.

Dual Learning in an Emergency Medicine Clerkship Improves Student Performance

BackgroundThe emergency department (ED) is an ideal environment to teach learners about the “undifferentiated patient.” Student learning may be inconsistent because of inherent variability in the ED. Previous research has suggested that standardizing the emergency medicine (EM) clerkship by implementing didactics and requiring students to see patients with particular chief complaints improves educational outcomes. ObjectiveTo compare knowledge acquisition after a new curriculum to the traditional curriculum. MethodsThis was a prospective, quasiexperimental study of senior medical students in an EM clerkship. Students were assigned to the dual learning (DL) group or standard learning (SL) groups based on month of rotation. All were required to see patients with 10 specific chief complaints and were lent an EM textbook. The SL group was instructed to read about the required cases. The DL group attended a 2-hour didactic session covering 5 of the 10 required cases. All students completed an identical pre- and postclerkship multiple choice knowledge test. ResultsData from 51 medical students (DL = 27; SL = 24) were analyzed. Mean pretest scores were comparable between groups. A 2 (groups) by 2 (sessions) mixed-design analysis of variance yielded a significant group by session interaction effect (p < 0.001). The DL group significantly increased its mean score from 8.7 (standard deviation [SD] = 1.8) pretest to 11.6 (SD = 1.9) posttest; there was no improvement in the SL group (pretest: 9.3 [SD = 1.5], posttest: 10.0 [SD = 2.0]). ConclusionA DL model combining clinical and enhanced didactic requirements for an EM clerkship led to greater knowledge gain than the standard curriculum. This model may suggest ways to improve the educational experience in the EM clerkship.