Leveraging enterprise knowledge graph to infer web events’ influences via self-supervised learning

Knowledge graph (KG) techniques have achieved successful results in many tasks, especially in semantic web and natural language processing domains. In recent years, representation learning on KG has been successfully applied to e-business applications, such as event-driven automatic investment strategies. However, there is still limited research about learning events’ influence on KG for modern quantitative investment. In this paper, we propose a novel event influence learning framework to predict stock market trends, called ST-Trend, leveraging enterprise knowledge graph to represent company correlation relationships, for mining the deep background knowledge of web events, with three self-supervised learning tasks. In particular, we devise two jointly self-supervised tasks to identify the relations between web events and companies. The first task is for generating ground-truth event-company correlation labels based on the enterprise knowledge graph. The second task is used to train how to identify the correlated companies of an event based on the generated correlation labels, with the encoding of web events, company features, and technical sequential data. We then design the prediction network to infer an event’s influence on stock price trends of the identified correlated companies based on the enterprise KG. Finally, we perform extensive experiments on a massive real-life dataset to validate the effectiveness of our proposed framework, and the experimental results demonstrate its superior performance in predicting stock market trends via considering events’ influences with the enterprise knowledge graph.     

Fault reconstruction for Takagi–Sugeno fuzzy systems via learning observers

This paper addresses the problem of observer-based fault reconstruction for Takagi–Sugeno fuzzy systems. Two types of fuzzy learning observers are constructed to achieve simultaneous reconstruction of system states and actuator faults. Stability and convergence of the proposed observers are proved using Lyapunov stability theory, and necessary conditions for the existence of the observers are further discussed. The design of fuzzy learning observers can be formulated in terms of a series of linear matrix inequalities that can be conveniently solved using convex optimisation technique. A single-link flexible manipulator is employed to verify the effectiveness of the proposed fault-reconstructing approaches.     

Simulation-based learning: The learning–forgetting–relearning process and impact of learning history

The results of empirical experiments evaluating the effectiveness and efficiency of the learning–forgetting–relearning process in a dynamic project management simulation environment are reported. Sixty-six graduate engineering students performed repetitive simulation-runs with a break period of several weeks between the runs. The students used a teaching tool called the project management trainer (PMT) that simulates a generic dynamic, stochastic project management environment. In this research, we focused on the effect of history recording mechanism on the learning forgetting process. Manual or automatic history recording mechanisms were used by the experimental group, while the control group did not use any history recording mechanism. The findings indicate that for the initial learning phase, the manual mechanism is better than the automatic mechanism. However, for the relearning phase, the break period length influenced the performance after the break. When the break period is short, the manual history keeping mechanism is better, but for a long period break, there is no significant difference. A comparison between the experimental group and the control group revealed that using any history recording mechanism reduced forgetting. Based on the findings, some practical implications of using simulators to improve the learning–forgetting process are discussed.     

Student teachers’ satisfaction for blended learning via Edmodo learning management system

In a digitally driven world, behaviours of future teachers for blended learning (both face-to-face and on-line classes) need to be examined. This study serves three purposes. The first is to examine student teachers’ preferences for Community-of-Inquiry model-driven blended learning via Edmodo. Second, predicting student satisfaction on b-learning from a combination of four variables (gender, having internet access, using the internet for information access, and previous experience in on-line learning) was questioned. And third, b-learning orientations of participants were investigated. One of the mixed methods, the concurrent triangulation design was employed in which both qualitative and quantitative methods were applied. The study group included 135 freshmen and junior students (29 males and 106 females) from a western Turkish educational faculty. The findings for the first question indicated that 70.4% of student teachers prefer b-learning. For the second, 15% of the variance in satisfaction on b-learning was explained by the proposed model with a medium effect. And for the third, the qualitative findings were discussed under Perceived Usefulness (PU) and Perceived Uselessness and Unease-of-Use (PU-UU) themes. Although less than a quarter of participants found b-learning useless, most held positive notions for b-learning practices via Edmodo.     

Synchronization of delayed reaction–diffusion neural networks via an adaptive learning control approach

An adaptive learning control strategy is utilized to investigate the synchronization problem for delayed reaction–diffusion neural networks (RDNNs) with unknown time-varying coupling strengths. A novel adaptive synchronization approach is proposed, which is consisted of differential–difference type updating law and feedback control law. By constructing a Lyapunov–Krasovskii-like composite energy functional (CEF), based on the LaSalle invariant principle of functional differential equations, a sufficient condition for the adaptive synchronization of such a system is obtained. Finally, a numerical example is given to show the effectiveness of the proposed synchronization method.     

Students’ multimodal knowledge practices in a makerspace learning environment

International Journal of Computer-Supported Collaborative Learning - In this study, we aim to widen the understanding of how students’ collaborative knowledge practices are mediated...     

Feature extraction using median–mean and feature line embedding

In this article, we propose a feature extraction method based on median–mean and feature line embedding (MMFLE) for the classification of hyperspectral images. In MMFLE, we maximize the class separability using discriminant analysis. Moreover, we remove the negative effect of outliers on the class mean using the median–mean line (MML) measurement and virtually enlarge the training set using the feature line (FL) distance metric. The experimental results on Indian Pines and University of Pavia data sets show the better performance of MMFLE compared to nearest feature line embedding (NFLE), median–mean line discriminant analysis (MMLDA), and some other feature extraction approaches in terms of classification accuracy using a small training set.     

Object–object interaction affordance learning

This paper presents a novel object–object affordance learning approach that enables intelligent robots to learn the interactive functionalities of objects from human demonstrations in everyday environments. Instead of considering a single object, we model the interactive motions between paired objects in a human–object–object way. The innate interaction-affordance knowledge of the paired objects are learned from a labeled training dataset that contains a set of relative motions of the paired objects, human actions, and object labels. The learned knowledge is represented with a Bayesian Network, and the network can be used to improve the recognition reliability of both objects and human actions and to generate proper manipulation motion for a robot if a pair of objects is recognized. This paper also presents an image-based visual servoing approach that uses the learned motion features of the affordance in interaction as the control goals to control a robot to perform manipulation tasks.     

Estimating variable structure and dependence in multitask learning via gradients

We consider the problem of hierarchical or multitask modeling where we simultaneously learn the regression function and the underlying geometry and dependence between variables. We demonstrate how the gradients of the multiple related regression functions over the tasks allow for dimension reduction and inference of dependencies across tasks jointly and for each task individually. We provide Tikhonov regularization algorithms for both classification and regression that are efficient and robust for high-dimensional data, and a mechanism for incorporating a priori knowledge of task (dis)similarity into this framework. The utility of this method is illustrated on simulated and real data.     

Event-triggered H∞ consensus control for input-constrained multi-agent systems via reinforcement learning

This article presents an event-triggered H∞ consensus control scheme using reinforcement learning (RL) for nonlinear second-order multi-agent systems (MASs) with control constraints. First, considering control constraints, the constrained H∞ consensus problem is transformed into a multi-player zero-sum game with non-quadratic performance functions. Then, an event-triggered control method is presented to conserve communication resources and a new triggering condition is developed for each agent to make the triggering threshold independent of the disturbance attenuation level. To derive the optimal controller that can minimize the cost function in the case of worst disturbance, a constrained Hamilton–Jacobi–Bellman (HJB) equation is defined. Since it is difficult to solve analytically due to its strongly non-linearity, reinforcement learning (RL) is implemented to obtain the optimal controller. In specific, the optimal performance function and the worst-case disturbance are approximated by a time-triggered critic network; meanwhile, the optimal controller is approximated by event-triggered actor network. After that, Lyapunov analysis is utilized to prove the uniformly ultimately bounded (UUB) stability of the system and that the network weight errors are UUB. Finally, a simulation example is utilized to demonstrate the effectiveness of the control strategy provided.     

On kernel engineering via Paley–Wiener

A radial basis function approximation takes the form

$s(x)=\sum_{k=1}^na_k\phi(x-b_k),\quad x\in {\mathbb{R}}^d,$

where the coefficients a ₁,…,a _n are real numbers, the centres b ₁,…,b _n are distinct points in ? ^d , and the function φ:? ^d →? is radially symmetric. Such functions are highly useful in practice and enjoy many beautiful theoretical properties. In particular, much work has been devoted to the polyharmonic radial basis functions, for which φ is the fundamental solution of some iterate of the Laplacian. In this note, we consider the construction of a rotation-invariant signed (Borel) measure μ for which the convolution ψ=μ φ is a function of compact support, and when φ is polyharmonic. The novelty of this construction is its use of the Paley–Wiener theorem to identify compact support via analysis of the Fourier transform of the new kernel ψ, so providing a new form of kernel engineering.

     

Inductive hierarchical nonnegative graph embedding for “verb–object” image classification

Most existing image classification algorithms mainly focus on dealing with images with only “object” concepts. However, in real-world cases, a great variety of images contain “verb–object” concepts, rather than only “object” ones. The hierarchical structure embedded in these “verb–object” concepts can help to enhance classification. However, traditional feature representation methods cannot utilize it. To tackle this problem, we present in this paper a novel approach, called inductive hierarchical nonnegative graph embedding. By assuming that those “verb–object” concept images which share the same “object” part but different “verb” part have a specific hierarchical structure, we integrate this hierarchical structure into the nonnegative graph embedding technique, together with the definition of inductive matrix, to (1) conduct effective feature extraction from hierarchical structure, (2) easily transfer each new testing sample into its low-dimensional nonnegative representation, and (3) perform image classification of “verb–object” concept images. Extensive experiments compared with the state-of-the-art algorithms on nonnegative data factorization demonstrate the classification power of proposed approach on “verb–object” concept images classification.     

Efficient behavior learning in human–robot collaboration

We present a novel method for a robot to interactively learn, while executing, a joint human–robot task. We consider collaborative tasks realized by a team of a human operator and a robot helper that adapts to the human’s task execution preferences. Different human operators can have different abilities, experiences, and personal preferences so that a particular allocation of activities in the team is preferred over another. Our main goal is to have the robot learn the task and the preferences of the user to provide a more efficient and acceptable joint task execution. We cast concurrent multi-agent collaboration as a semi-Markov decision process and show how to model the team behavior and learn the expected robot behavior. We further propose an interactive learning framework and we evaluate it both in simulation and on a real robotic setup to show the system can effectively learn and adapt to human expectations.     

QSIA – a Web-based environment for learning,assessing and knowledge sharing in communities

This paper describes a Web-based and distributed system named QSIA that serves as an environment for learning, assessing and knowledge sharing. QSIA – Questions Sharing and Interactive Assignments – offers a unified infrastructure for developing, collecting, managing and sharing of knowledge items. QSIA enhances collaboration in authoring via online recommendations and generates communities of teachers and learners. At the same time, QSIA fosters individual learning and might promote high-order thinking skills among its users. QSIA's community, conceptual architecture, structure overview and implementations are discussed.     

Enhancing students’ knowledge building through utilising social interactions in an online learning environment

ABSTRACT

In the new “open world” of information, educational systems should involve students in constructing new knowledge of value to a community out of fragmentary information. The already proposed Knowledge Building (KB) approaches typically support only a few general-purpose activities due to the constraints of the utilised web-based environments. To organise and facilitate students’ KB during course activities, this study incorporated services provided by DoosMooc social learning environment into a knowledge transformation model. This approach is completely adapted to an educational context and allows time for iterations, helping students to both contribute to social KB processes and take collective responsibility for improving their understanding of authentic problems. The features provided by the introduced environment support and assess students’ KB activities and facilitate processes of creating, representing, organising, and reviewing different types of knowledge artefacts. The results of a semester-long experiment indicate that the approach and the corresponding instructional design thereof could successfully organise students’ KB activities and facilitate the required interactions. This study reports the impacts of parameters such as learner expertise and quality of shared knowledge on the planned KB processes, and investigates the relationships between students' KB activities and learning achievements.     

An exploration of the uncertainty relation satisfied by BP network learning ability and generalization ability

An overfit phenomenon exists in the BP network. The so-called overfit means that as long as the network is allowed to be sufficiently complicated, the BP network can minimize the error of the training sample set; however, in the case of a limited number of samples, the generalization ability of the network will decrease. This indicates that there is a relation between the learning ability and the generalization ability. Therefore, studying the relationship between the learning ability is the…     

Design of delta–sigma modulators via generalized Kalman–Yakubovich–Popov lemma

This paper is concerned with the design of delta–sigma modulators via the generalized Kalman–Yakubovich–Popov lemma. The shaped noise transfer function (NTF) is assumed to have infinite impulse response, and the optimization objective is minimizing the maximum magnitude of the NTF over the signal frequency band. By virtue of the GKYP lemma, the optimization of an NTF is converted into a minimization problem subject to quadratic matrix inequalities, and then an iterative algorithm is proposed to solve this alternative minimization problem. Each iteration of the algorithm contains linear matrix inequality constraints only and can be effectively solved by the existing numerical software packages. Moreover, specifications on the NTF zeros are also integrated in the design method. A design example demonstrates that the proposed design method has an advantage over the benchmark one in improving the signal-to-noise ratio.     

Joint input–output identification of unstable systems with kernel regularization

This paper discusses closed-loop identification of unstable systems. In particular, we first apply the joint input–output identification method and then convert the identification problem of unstable systems into that of stable systems, which can be tackled by using kernel-based regularization methods.We propose to identify two transfer functions by kernel regularization, the one from the reference signal to the input, and the one from the reference signal to the output. Since these transfer functions are stable, kernel regularization methods can construct their accurate models. Then the model of unstable system is constructed by ratio of these functions. The effectiveness of the proposed method is demonstrated by a numerical example and a practical experiment with a DC motor.     

Reinforcement learning technique using agent state occurrence frequency with analysis of knowledge sharing on the agent’s learning process in multiagent environments

Reinforcement learning techniques like the Q-Learning one as well as the Multiple-Lookahead-Levels one that we introduced in our prior work require the agent to complete an initial exploratory path followed by as many hypothetical and physical paths as necessary to find the optimal path to the goal. This paper introduces a reinforcement learning technique that uses a distance measure to the goal as a primary gauge for an autonomous agent’s action selection. In this paper, we take advantage of the first random walk to acquire initial information about the goal. Once the agent’s goal is reached, the agent’s first perceived internal model of the environment is updated to reflect and include said goal. This is done by the agent tracing back its steps to its origin starting point. We show in this paper, no exploratory or hypothetical paths are required after the goal is initially reached or detected, and the agent requires a maximum of two physical paths to find the optimal path to the goal. The agent’s state occurrence frequency is introduced as well and used to support the proposed Distance-Only technique. A computation speed performance analysis is carried out, and the Distance-and-Frequency technique is shown to require less computation time than the Q-Learning one. Furthermore, we present and demonstrate how multiple agents using the Distance-and-Frequency technique can share knowledge of the environment and study the effect of that knowledge sharing on the agents’ learning process.