Learning Problems Research Articles

Problems of Distance Learning in EU Countries During the COVID-19 Pandemic

The article is dedicated to the analysis of distance learning problems faced by EU Member states during the COVID-19 pandemic. Each EU member state was forced to transform its education policy, trying not only to reduce the risks of infection during professional contacts between teachers and students, but also to ensure the maintenance of the necessary quality of the educational process. The ambiguous influence of distance learning on the level of academic achievement of students in various EU countries is shown. The factors that hindered distance learning, including low incomes, housing and other material deprivation, belonging of students and their families to socially vulnerable segments of the population, the level of education of parents of schoolchildren and students are identified. The negative consequences of restrictive measures implemented by the EU countries in the education sector are revealed. Attention is focused on the risk of reducing the level of competitiveness of future graduates of educational institutions in the labor market due to their lack of knowledge. The reaction of the national authorities of the EU countries to the situation in the education sector, expressed in the adoption of a set of urgent measures, is investigated, including additional classes with lagging students, the preparation and active use of teaching materials, the provision of free SIM cards for an unlimited period of time to students from low-income families, providing computers and Internet access to children in need. It is stated that the education sector in the EU countries has demonstrated the necessary adaptability to the situation, having found support from the authorities.

Margin-aware Noise-robust Contrastive Learning for Partially View-aligned Problem

In this paper, we study a challenging problem in contrastive learning when just a portion of data is aligned in multi-view dataset due to temporal, spatial, or spatio-temporal asynchronism across views. It is important to study partially view-aligned data since this type of data is common in real-world application and easily leads to data inconsistency among different views. Such a Partially View-aligned Problem (PVP) in contrastive learning has been relatively less touched so far, especially in downstream tasks, i.e., classification and clustering. In order to solve this problem, we introduce a flexible margin and propose margin-aware noise-robust contrastive learning to simultaneously identify the within-category counterparts from the other view of one data point based on the established cross-view correspondence and learn a shared representation. To be specific, the proposed learning framework is built on a novel margin-aware noise-robust contrastive loss. Since data pairs are used as input for the proposed margin-aware noise-robust contrastive learning, we build positive pairs according to the known correspondences and negative pairs in the manner of random sampling. Our margin-aware noise-robust contrastive learning framework is able to effectively reduce or remove the impacts caused by the possible existing noise for the constructed pairs in a margin-aware manner, i.e., false negative pairs leaded by random sampling in PVP. We relax the proposed margin-aware noise-robust contrastive loss and then give a detailed mathematical analysis for the effectiveness of our loss. As an instantiation, we construct an example under the proposed margin-aware noise-robust contrastive learning framework for validation in this work. To the best of our knowledge, this is the first attempt of extending contrastive learning to a margin-aware noise-robust version for dealing with PVP. We also enrich the learning paradigm when there is noise in the data. Extensive experiments on different datasets demonstrate the promising performance of the proposed method in the classification and clustering tasks.

ℓp-Sphere Covering and Approximating Nuclear p-Norm

The spectral p-norm and nuclear p-norm of matrices and tensors appear in various applications, albeit both are NP-hard to compute. The former sets a foundation of [Formula: see text]-sphere-constrained polynomial optimization problems, and the latter has been found in many rank minimization problems in machine learning. We study approximation algorithms of the tensor nuclear p-norm with an aim to establish the approximation bound matching the best one of its dual norm, the tensor spectral p-norm. Driven by the application of sphere covering to approximate both tensor spectral and nuclear norms ([Formula: see text]), we propose several types of hitting sets that approximately represent the [Formula: see text]-sphere with adjustable parameters for different levels of approximations and cardinalities, providing an independent toolbox for decision making on [Formula: see text]-spheres. Using the idea in robust optimization and second-order cone programming, we obtain the first polynomial-time algorithm with an [Formula: see text]-approximation bound for the computation of the matrix nuclear p-norm when [Formula: see text] is a rational, paving a way for applications in modeling with the matrix nuclear p-norm. These two new results enable us to propose various polynomial-time approximation algorithms for the computation of the tensor nuclear p-norm using tensor partitions, convex optimization, and duality theory, attaining the same approximation bound to the best one of the tensor spectral p-norms. Effective performance of the proposed algorithms for the tensor nuclear p-norm is shown by numerical implementations. We believe the ideas of [Formula: see text]-sphere covering with its applications in approximating nuclear p-norm would be useful for tackling optimization problems on other sets, such as the binary hypercube with its applications in graph theory and neural networks and the nonnegative sphere with its applications in copositive programming and nonnegative matrix factorization. Funding: This work was supported by the National Natural Science Foundation of China [Grants 71825003, 72394360, 72394364, 72171141, 72192830, and 72192832], and the program for Innovative Research Team of Shanghai University of Finance and Economics.

Credible Negotiation for Multi-agent Reinforcement Learning in Long-term Coordination

The coordination of multi-agent is one of the critical problems in Multi-Agent Reinforcement Learning (MARL). The traditional methods of MARL focus on finding a stochastically acceptable solution called Nash Equilibrium (NE) for all agents from the Markov game in which multiple equilibria exist. However, learning a fair equilibrium is crucial for the sustainability and stability of collaboration in the long-term coordination game, especially when the leadership competition exists. In this paper, we propose the bi-level reinforcement learning method N-Bi-AC, whose solution is a Pareto improvement for traditional NE, to choose a fair Equilibrium. There are two parts in our method, the first is that we propose the Negotiator to determine the leader in stage game, and the other is to update the Q-value of agents in the game by using a bi-level actor-critic learning method based on the Joint Mixed Strategy Equilibrium Q-learning algorithm (JMSE Q-learning). The convergence proof is given, and the learning algorithm is compared with the state-of-the-art algorithms. We found that the proposed N-Bi-AC method successfully converged to a fair Nash Equilibrium, which guarantees the fairness of agents in different matrix game environments.

Research on the Problems in Chinese Classroom Teaching in Ethnic Regions—Taking Four Primary Schools in Nagqu City as an Example

Classroom teaching is the main channel of Chinese language teaching, which is the process of imparting Chinese knowledge, cultivating Chinese core literacy and abilities through teaching activities in the classroom. By means of interviews and classroom observation, found that in the classroom teaching field, western China’s primary school unified language textbooks have been put into use since the autumn of 2017, and have had problems in teacher teaching, student learning, curriculum nature and classroom culture. Based on this, it is proposed that putting literacy first, emphasizing teaching research, and striving for teaching effectiveness; Promoting student development through mutual aid in lesson observation and situational teaching; Based on text, combining reading and writing to enjoy the beauty of Chinese language; Integrating education into culture, creating an infectious atmosphere, and showcasing the art of teaching, with a view to improving the quality of Chinese education and teaching in western China.

Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models.

In deep learning, different kinds of deep networks typically need different optimizers, which have to be chosen after multiple trials, making the training process inefficient. To relieve this issue and consistently improve the model training speed across deep networks, we propose the ADAptive Nesterov momentum algorithm, Adan for short. Adan first reformulates the vanilla Nesterov acceleration to develop a new Nesterov momentum estimation (NME) method, which avoids the extra overhead of computing gradient at the extrapolation point. Then Adan adopts NME to estimate the gradient's first- and second-order moments in adaptive gradient algorithms for convergence acceleration. Besides, we prove that Adan finds an ϵ-approximate first-order stationary point within O(ϵ-3.5) stochastic gradient complexity on the non-convex stochastic problems (e.g., deep learning problems), matching the best-known lower bound. Extensive experimental results show that Adan consistently surpasses the corresponding SoTA optimizers on vision, language, and RL tasks and sets new SoTAs for many popular networks and frameworks, e.g., ResNet, ConvNext, ViT, Swin, MAE, DETR, GPT-2, Transformer-XL, and BERT. More surprisingly, Adan can use half of the training cost (epochs) of SoTA optimizers to achieve higher or comparable performance on ViT, GPT-2, MAE, etc, and also shows great tolerance to a large range of minibatch size, e.g., from 1 k to 32 k.

Adaptive network approach to exploration-exploitation trade-off in reinforcement learning.

A foundational machine-learning architecture is reinforcement learning, where an outstanding problem is achieving an optimal balance between exploration and exploitation. Specifically, exploration enables the agents to discover optimal policies in unknown domains of the environment for gaining potentially large future rewards, while exploitation relies on the already acquired knowledge to maximize the immediate rewards. We articulate an approach to this problem, treating the dynamical process of reinforcement learning as a Markov decision process that can be modeled as a nondeterministic finite automaton and defining a subset of states in the automaton to represent the preference for exploring unknown domains of the environment. Exploration is prioritized by assigning higher transition probabilities to these states. We derive a mathematical framework to systematically balance exploration and exploitation by formulating it as a mixed integer programming (MIP) problem to optimize the agent's actions and maximize the discovery of novel preferential states. Solving the MIP problem provides a trade-off point between exploiting known states and exploring unexplored regions. We validate the framework computationally with a benchmark system and argue that the articulated automaton is effectively an adaptive network with a time-varying connection matrix, where the states in the automaton are nodes and the transitions among the states represent the edges. The network is adaptive because the transition probabilities evolve over time. The established connection between the adaptive automaton arising from reinforcement learning and the adaptive network opens the door to applying theories of complex dynamical networks to address frontier problems in machine learning and artificial intelligence.

Object-Centric Representation Learning for Video Scene Understanding.

Depth-aware Video Panoptic Segmentation (DVPS) is a challenging task that requires predicting the semantic class and 3D depth of each pixel in a video, while also segmenting and consistently tracking objects across frames. Predominant methodologies treat this as a multi-task learning problem, tackling each constituent task independently, thus restricting their capacity to leverage interrelationships amongst tasks and requiring parameter tuning for each task. To surmount these constraints, we present Slot-IVPS, a new approach employing an object-centric model to acquire unified object representations, thereby facilitating the model's ability to simultaneously capture semantic and depth information. Specifically, we introduce a novel representation, Integrated Panoptic Slots (IPS), to capture both semantic and depth information for all panoptic objects within a video, encompassing background semantics and foreground instances. Subsequently, we propose an integrated feature generator and enhancer to extract depth-aware features, alongside the Integrated Video Panoptic Retriever (IVPR), which iteratively retrieves spatial-temporal coherent object features and encodes them into IPS. The resulting IPS can be effortlessly decoded into an array of video outputs, including depth maps, classifications, masks, and object instance IDs. We undertake comprehensive analyses across four datasets, attaining state-of-the-art performance in both Depth-aware Video Panoptic Segmentation and Video Panoptic Segmentation tasks.

Approaching the Global Nash Equilibrium of Non-Convex Multi-Player Games.

Many machine learning problems can be formulated as non-convex multi-player games. Due to non-convexity, it is challenging to obtain the existence condition of the global Nash equilibrium (NE) and design theoretically guaranteed algorithms. This paper studies a class of non-convex multi-player games, where players' payoff functions consist of canonical functions and quadratic operators. We leverage conjugate properties to transform the complementary problem into a variational inequality (VI) problem using a continuous pseudo-gradient mapping. We prove the existence condition of the global NE as the solution to the VI problem satisfies a duality relation. We then design an ordinary differential equation to approach the global NE with an exponential convergence rate. For practical implementation, we derive a discretized algorithm and apply it to two scenarios: multi-player games with generalized monotonicity and multi-player potential games. In the two settings, step sizes are required to be O(1/k) and O(1/√k) to yield the convergence rates of O(1/ k) and O(1/√k), respectively. Extensive experiments on robust neural network training and sensor network localization validate our theory. Our code is available at https://github.com/GuanpuChen/Global-NE.

Uncertainty-Boosted Robust Video Activity Anticipation.

Video activity anticipation aims to predict what will happen in the future, embracing a broad application prospect ranging from robot vision and autonomous driving. Despite the recent progress, the data uncertainty issue, reflected as the content evolution process and dynamic correlation in event labels, has been somehow ignored. This reduces the model generalization ability and deep understanding on video content, leading to serious error accumulation and degraded performance. In this paper, we address the uncertainty learning problem and propose an uncertainty-boosted robust video activity anticipation framework, which generates uncertainty values to indicate the credibility of the anticipation results. The uncertainty value is used to derive a temperature parameter in the softmax function to modulate the predicted target activity distribution. To guarantee the distribution adjustment, we construct a reasonable target activity label representation by incorporating the activity evolution from the temporal class correlation and the semantic relationship. Moreover, we quantify the uncertainty into relative values by comparing the uncertainty among sample pairs and their temporal-lengths. This relative strategy provides a more accessible way in uncertainty modeling than quantifying the absolute uncertainty values on the whole dataset. Experiments on multiple backbones and benchmarks show our framework achieves promising performance and better robustness/interpretability.

Trauma from cyberbullying: How it affects learning impairments

Investigating the effects of the traumatic experiences that adolescents with learning problems go through is the goal of our study. 160 high school students who were referred via a web-based evaluation form because of poor academic performance made up the sample for our study. In particular, the subscales for anxiety-depression and depression-withdrawal were employed in the ACHENBACH self-report questionnaire for children and teenagers. Calculations have been made for frequencies, response percentages, and means. Additionally, a one-way anova was used to conduct an analysis of variance to compare the average student responses to the variable "experiencing a traumatic event" between groups of students. The findings demonstrated that there are substantial variations in the means of students who claim having experienced a traumatic event and those who report not having, with a statistical significance threshold of p 0,01 between the two groups. Students' exposure to violence in schools and the loss of a loved one appear to be particularly significant traumatic experiences for their psyche. Additionally, a quick presentation of the application's cyberbullying-related capabilities is made.

A Comprehensive Analysis of Optimization Algorithms for Large-Scale Machine Learning

Abstract—Optimization algorithms determine the best model parameters that tend to offer prediction accuracy in complex machine learning (ML) problems in big data. In fact, they are very crucial while handling large-scale data because old ways of handling datasets will be computationally very inefficient. This paper provides an in-depth comparison of the two most widely used optimization techniques in machine learning: SGD, and BGD. Our simulated results indicate that though the SGD technique indicates very fast initial convergence, its efficiency eventually tends to degrade with the increasing iteration of the algorithm. On the other hand, the BGD approach might take slow initiation but is relatively consistent in a long run. We then further probe how the variations in learning rate affect its performance in the case of both the methods. Our analysis shows that adaptive learning rates drastically accelerate convergence. Finally, we show that the computational efficiency of the SGD method makes it a better choice since gradients can be computed on a per sample basis, which makes the method better for scaling. Keywords—Optimization Algorithms, Stochastic gradient de- scent, Batch gradient descent, Machine learning, Large-scale data, and Adaptive learning rate.

Virtual Reality in Algorithm Programming Course: Practicality and Implications for College Students

Virtual Reality (VR) is one of the revolutionary technologies in education that can simulate learning interactively and realistically. The reliability of VR supports various variations in learning, including learning programming algorithms. This research aims to design and develop VR in learning programming algorithms. The ADDIE model is applied in this study to develop products through the stages of analysis, design, development, implementation, and evaluation. The analysis of learning problems shows the unavailability of interactive learning media that can support various learning styles of students in programming algorithm materials. Therefore, this VR learning design integrates various elements of students' learning styles, including visual, audiovisual, kinesthetic, and reading. The product developed is in the form of a VR application that presents learning by having a variety of learning styles. Audiovisual displays are presented in a VR environment, while kinesthetic interactions are obtained through user movements, and material information and readings are presented in a VR environment. The data collected included the results of a practicality and effectiveness test involving sixty-six students. The results of the practicality test show that this VR product is practical to use based on the indicators of convenience, attractiveness, efficiency, and benefits. Meanwhile, the effectiveness test shows that this VR product can improve learning outcomes better than traditional learning methods. Overall, the VR products developed have proven to be practical and effective in learning programming algorithms. The presence of VR, which presents a variety of realistic displays in a virtual environment, opens opportunities for further research related to distance learning through VR technology as well as an analysis of the impact of VR on health.

Few-Shot Learning Method Based on Data Enhancement and Transfer Learning

With the significant progress of deep learning technology in many fields, the dependence of model training on a large amount of labeled data is increasingly prominent. However, in many practical application scenarios, especially in tasks with high labeling costs, data scarcity often occurs. This realistic challenge has promoted the rise of Few-Shot Learning (FSL) technology, which seeks to achieve effective learning of models with extremely limited samples. This article provides a comprehensive overview of the theoretical background, key technologies of FSL, to explore its potential and effectiveness in solving the problem of small-sample learning. In the method overview section, this article pays special attention to FSL strategies based on data augmentation and transfer learning. By reviewing and analyzing these methods, this article aims to provide some theoretical support and technical references for further exploration in this field and hopes to contribute to solving the problem of data scarcity and promote the sustainable development of this field.

Analysis on Problems and Strategies of Web-based Autonomous Learning in English Courses of Open Education

With the rapid development of network technology, it puts forward greater requirements on autonomous learning ability in English courses of open education. Based on the problems in autonomous learning in English courses of open education, the thesis analyses the developmental strategies of enhancing autonomous learning abilities in English online courses of open education.

Pengembangan Multimedia Pembelajaran IPA pada Materi Spesialisasi Sel Kelas VIII SMP

The development of this multimedia learning is motivated by the problem of learning, where teachers only use printed books as the main learning media in the classroom. Students tend to feel bored and less interested in following lessons using printed books that only contain text and images. This research aims to produce valid and practical multimedia learning products that can be used in the learning process. The type of research used is research and development, also known as Research and Development (R&D). The development model used is the 4D model, which consists of four main stages: define, design, development, and disseminate. The product validity test was conducted by 4 validators, namely 2 material validators and 2 media validators. The product trial was conducted on 20 eighth-grade students of SMP Negeri 5 Linggo Sari Baganti with the aim of determining the practicality of the developed product. The results of the material validation by 2 material validators obtained an average score of 4.93 with the category "very valid." The results from the assessment by 2 media validators obtained an average score of 4.45 with the category "very valid." The results of the media practicality test obtained an average score of 4.45 with the category "very practical." Based on the results of the validity and practicality tests, it can be concluded that the IPA learning multimedia on the "Cell Specialization" material for 8th-grade junior high school produced is valid and practical, thus it can be used in the learning process

COMPARATIVE ANALYSIS OF AUTOMATED MACHINE LEARNING LIBRARIES: PYCARET, H2O, TPOT, AUTO-SKLEARN, AND FLAML

Automated Machine Learning (AutoML) frameworks have gained significant popularity in recent years, making machine learning accessible to a broader audience by automating many of the tasks traditionally performed by data scientists and domain experts. This paper presents a comparative analysis of five prominent AutoML libraries: PyCaret, H2O.ai, TPOT, Auto-sklearn, and FLAML. Each of these libraries provides automated solutions for model selection, hyperparameter tuning, and other machine learning tasks. The goal of this study is to assess their performance, ease of use, flexibility, and suitability for various types of machine learning problems. The comparison is based on multiple factors, including ease of use, performance, customization, resource efficiency, and suitability. This research aims to help researchers, practitioners, and developers in selecting the most appropriate AutoML library based on their specific needs and resources. Keywords: AutoML, PyCaret, H2O.ai, TPOT, Auto-sklearn, FLAML

Teacher training and multidimensional systems management for the delivery of mental health care for schools

ABSTRACT Objectives According to the World Health Organisation (WHO), more than one in four young people globally experience a mental health issue each year. Alongside traditional care undertaken by hospitals, the justice system, and community care, schools also provide care for students with mental health issues. As a result of increased numbers of students presenting in school with undiagnosed disorders and behavioural and learning problems, schoolteachers have been asking for more mental health training. We sought to explore teachers’ needs for a mental health teacher-training curriculum to support students with mental health issues. Method A qualitative method was used consisting of 18 in-depth interviews with currently serving teachers from Catholic, Independent and Public schools in Australia. Data were analysed using thematic content analysis. Results Data revealed that the immediate requirement for training is knowledge of mental health. However, we found this to be insufficient to support teachers and found instead that a multidimensional model was needed comprising of training in the following areas: mental health management skills; training to deliver mental health programs in class; working with and partnering with external agencies, including roles and referral protocolsls; as well as caring teachers own mental health. Discussion We developed a multidimensional model for teacher training and for other school staff to meet the increasing demand on schools to manage students’ mental health issues in their classrooms.

Quantum Circuit Architecture Search on a Superconducting Processor

Variational quantum algorithms (VQAs) have shown strong evidence to gain provable computational advantages in diverse fields such as finance, machine learning, and chemistry. However, the heuristic ansatz exploited in modern VQAs is incapable of balancing the trade-off between expressivity and trainability, which may lead to degraded performance when executed on noisy intermediate-scale quantum (NISQ) machines. To address this issue, here, we demonstrate the first proof-of-principle experiment of applying an efficient automatic ansatz design technique, i.e., quantum architecture search (QAS), to enhance VQAs on an 8-qubit superconducting quantum processor. In particular, we apply QAS to tailor the hardware-efficient ansatz toward classification tasks. Compared with heuristic ansätze, the ansatz designed by QAS improves the test accuracy from 31% to 98%. We further explain this superior performance by visualizing the loss landscape and analyzing effective parameters of all ansätze. Our work provides concrete guidance for developing variable ansätze to tackle various large-scale quantum learning problems with advantages.

Adaptive pruning-based Newton's method for distributed learning

Newton's method leverages curvature information to boost performance, and thus outperforms first-order methods for distributed learning problems. However, Newton's method is not practical in large-scale and heterogeneous learning environments, due to obstacles such as high computation and communication costs of the Hessian matrix, sub-model diversity, staleness of training, and data heterogeneity. To overcome these obstacles, this paper presents a novel and efficient algorithm named Distributed Adaptive Newton Learning (DANL), which solves the drawbacks of Newton's method by using a simple Hessian initialization and adaptive allocation of training regions. The algorithm exhibits remarkable convergence properties, which are rigorously examined under standard assumptions in stochastic optimization. The theoretical analysis proves that DANL attains a linear convergence rate while efficiently adapting to available resources and keeping high efficiency. Furthermore, DANL shows notable independence from the condition number of the problem and removes the necessity for complex parameter tuning. Experiments demonstrate that DANL achieves linear convergence with efficient communication and strong performance across different datasets.