Freely Moving Visuomotor Learning Assay with Reversal Learning v2

The goal of this protocol is to assess visuomotor learning and motor flexibility in freely-moving mice, using the Visiomode touchscreen platform. It modifies the original protocol's (dx.doi.org/10.17504/protocols.io.bumgnu3w) last stage by replacing forelimb reaching with a reversal learning paradigm

Integrating 21st Century Competencies Into a Digital Teaching and Learning Model

Globalization and technology are two features of the modern world impacting all activity, and the resultant effect on education is causing much to be questioned about the teaching and learning paradigm. Ways in which the learning experience must change in response to changing global demands placed on societies and economies forms a large part of the current discourse around reforming education. This chapter puts forward a definition of globalization, 21st century skills, and the four main competencies known as the 4Cs, and then makes links between these phenomena as a way of understanding the digitization of education. The connections are possible through a process of gathering reflections and experiences from experienced educational practitioners.

Analysis of an Online English Teaching Model Application Based on Improved Multiorganizational Particle Population Optimization Algorithm

This paper uses an improved multiorganizational particle population optimization algorithm to conduct an in-depth analysis and study of an online English teaching model and uses the altered model for practical applications. The model building elements are extracted from it for the initial construction of a blended learning model of English-speaking teaching in junior high school. The main purpose of the first round of action research is to test the rationality of each element of the model, the main purpose of the second round of action research is to refine the model links and improve the operability of the model, and the main purpose in the third round of action research is to test the perfected model and explore the model. The main purpose of the third round of action research is to test the refined model and explore the application suggestions of the model. After the three rounds of action research, we finally obtained a more mature blended learning model for teaching English as a foreign language in junior high school. Mainly through the comparison of the pre- and posttest scores of English speaking of the experimental subjects and the comparison of the pre- and posttest data of the relevant questionnaires, the following experimental conclusions were drawn: adopting the blended learning-based speaking teaching model can effectively improve students’ interest in learning English, their attitudes and their English speaking skills including pronunciation, phonetic intonation, conversational communication, and oral expression and can enhance students’ group cooperation and communication ability, independent learning ability, evaluation awareness, and ability. This single-guided learning mechanism can effectively avoid the shocks that are easily caused by the dual-guided role of traditional PSO. The dimensional learning strategy constructs a learning paradigm for each particle by learning from each dimension of the individual optimal position of the particle to the corresponding dimension of the group optimal position, respectively. Dimensional learning is formally integrated into the learning paradigm only if it can improve the fitness of the paradigm so that the dimensional learning strategy can avoid the phenomenon of degradation of the learning paradigm and the phenomenon of “two steps forward, one step back.” In the dimensional learning strategy, since each particle learns from best, although it has a strong exploitation capability, it may cause all particles to converge to best quickly, making the algorithm converge prematurely.

Attention-Based Graph Convolutional Network for Zero-Shot Learning with Pre-Training

Zero-shot learning (ZSL) is a powerful and promising learning paradigm for classifying instances that have not been seen in training. Although graph convolutional networks (GCNs) have recently shown great potential for the ZSL tasks, these models cannot adjust the constant connection weights between the nodes in knowledge graph and the neighbor nodes contribute equally to classify the central node. In this study, we apply an attention mechanism to adjust the connection weights adaptively to learn more important information for classifying unseen target nodes. First, we propose an attention graph convolutional network for zero-shot learning (AGCNZ) by integrating the attention mechanism and GCN directly. Then, in order to prevent the dilution of knowledge from distant nodes, we apply the dense graph propagation (DGP) model for the ZSL tasks and propose an attention dense graph propagation model for zero-shot learning (ADGPZ). Finally, we propose a modified loss function with a relaxation factor to further improve the performance of the learned classifier. Experimental results under different pre-training settings verified the effectiveness of the proposed attention-based models for ZSL.