Instant Learning Research Articles

Information Search Within a Web Page: Modeling the Full Sequence of Eye Movement Decisions, Subjective Value Updating, and First Clicks

Online retail settings often present shoppers with large, complex choice sets where they need to quickly and dynamically weigh the benefits and costs of search within each web page. We build a model of information search within a web page using eye-tracking data collected during two incentive-compatible online shopping experiments, in which participants browsed the websites of two different clothing retailers (Experiments 1 and 2), as well as previously reported data from a laboratory experiment involving choices among snack food assortments (Experiment 3). Our model incorporates features that build upon recent advances in descriptive and normative models of information sampling and search in psychology and economics. First, our model captures how people decide where to look by treating eye fixations on clickable options as a series of “split-second” decisions that depend on estimates of option attractiveness and navigation effort. Second, our model assumes that the value of each option is learned via Bayesian updating. Third, the choice to end search on the web page depends on a dynamic decision threshold. Our model outperforms benchmarks that assume random search, instant learning, fixed thresholds, nonheterogeneous thresholds, and stochastic accumulator stopping rules. Explicitly modeling the sequence of eye fixation decisions results in accurate counterfactual simulations of the effects of hypothetical product orderings on search duration and quality as verified using experimental manipulation, and it can be applied flexibly to a wide range of web-page layouts. Systematic differences across experiments highlight the importance of accounting for product familiarity, choice-set size, and the role of category outside options. This paper was accepted by Ilia Tsetlin, behavioral economics and decision analysis. Funding: This work was supported by the Jay H. Baker Retailing Center and the Wharton Behavioral Lab at the University of Pennsylvania. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.02983 .

IHELP: a model for instant learning of video coding in VR/AR real-time applications

iHELP: a model for instant learning of video coding in VR/AR real-time applications

Time-Sensitive Learning for Heterogeneous Federated Edge Intelligence

Real-time machine learning (ML) has recently attracted significant interest due to its potential to support instantaneous learning, adaptation, and decision making in a wide range of application domains, including self-driving vehicles, intelligent transportation, and industry automation. In this paper, we investigate real-time ML in a federated edge intelligence (FEI) system, an edge computing system that implements federated learning (FL) solutions based on data samples collected and uploaded from decentralized data networks, e.g., Internet-of-Things (IoT) and/or wireless sensor networks. FEI systems often exhibit heterogenous communication and computational resource distribution, as well as non-i.i.d. data samples arrived at different edge servers, resulting in long model training time and inefficient resource utilization. Motivated by this fact, we propose a time-sensitive federated learning (TS-FL) framework to minimize the overall run-time for collaboratively training a shared ML model with desirable accuracy. Training acceleration solutions for both TS-FL with synchronous coordination (TS-FL-SC) and asynchronous coordination (TS-FL-ASC) are investigated. To address the straggler effect in TS-FL-SC, we develop an analytical solution to characterize the impact of selecting different subsets of edge servers on the overall model training time. A server dropping-based solution is proposed to allow some slow-performance edge servers to be removed from participating in the model training if their impact on the resulting model accuracy is limited. A joint optimization algorithm is proposed to minimize the overall time consumption of model training by selecting participating edge servers, the local epoch number (the number of model training iterations per coordination), and the data batch size (the number of data samples for each model training iteration). Motivated by the fact that data samples at the slowest edge server may exhibit special characteristics that cannot be removed from model training, we develop an analytical expression to characterize the impact of both staleness effect of asynchronous coordination and straggler effect of FL on the time consumption of TS-FL-ASC. We propose a load forwarding-based solution that allows a slow edge server to offload part of its training samples to trusted edge servers with higher processing capability. We develop a hardware prototype to evaluate the model training time of a heterogeneous FEI system. Experimental results show that our proposed TS-FL-SC and TS-FL-ASC can provide up to 63% and 28% of reduction, in the overall model training time, respectively, compared with traditional FL solutions.

A Nonlinear PID-Incorporated Adaptive Stochastic Gradient Descent Algorithm for Latent Factor Analysis

High-dimensional and incomplete (HDI) data are commonly encountered in various big data-related applications concerning the complex interactions among numerous nodes, such as the user-item iterations in a recommender system. A stochastic gradient descent (SGD)-based latent factor analysis (LFA) model can perform efficient representation learning to such HDI data, thereby extracting useful knowledge from them. However, a standard SGD algorithm updates a latent factor based on the current stochastic gradient only, without the considerations on the past information, making a resultant model suffer from slow convergence. To address this critical issue, this paper proposes an Adaptive Non-linear PID-incorporated SGD (ANPS) algorithm with two-fold ideas: 1) rebuilding the instant learning error when computing the stochastic gradient following the principle of a nonlinear PID controller to incorporate past update information into the learning scheme efficiently, and 2) implementing gain parameter adaptation following the principle of particle swarm optimization (PSO). Experiments on six widely-adopted HDI datasets demonstrate that compared with state-of-the-art LFA models, an ANPS-based LFA model achieves significant advantage in both efficiency and accuracy. Moreover, its flexible gain parameter adaptation mechanism greatly boosts its practicability for real issues. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —In many industrial applications like recommender systems, social network systems, and cloud service systems, people usually encounter numerous nodes and their highly-incomplete relationships. An HDI matrix is commonly adopted to describe such specific relationships. One of the major challenges is to acquire useful knowledge from an HDI matrix efficiently and accurately for various data analysis tasks, e.g., accurate recommendation, community detection, and web service selection. An SGD-based LFA model has been widely adopted to tackle this issue. However, it suffers from slow convergence that leads to considerable time cost on large-scale datasets. This study proposes an ANPS algorithm following the principle of a nonlinear PID controller. With it, an ANPS-based LFA model is achieved, which possesses fast convergence rate on an industrial HDI matrix. The proposed ANPS algorithm can be leveraged for different types of various machine learning models, thereby improving their utility and scalability in practice.

Instant learning based on deep neural network with linear discriminant analysis features extraction for accurate iris recognition system

Instant learning based on deep neural network with linear discriminant analysis features extraction for accurate iris recognition system

Learning in a crisis moment: a randomized controlled trial in emergency bystander intervention

BackgroundOpioid overdose is the leading cause of injury-related death in the United States. Individuals who overdose outside of clinical settings have more positive clinical outcomes when they receive naloxone, an opioid antagonist, from a bystander as an early intervention before emergency personnel arrive. However, there is a gap in knowledge about successful instantaneous learning and intervention in a real-life stressful environment. The objective of this study is to explore the efficacy of different instructional delivery methods for bystanders in a stressful environment. We aim to evaluate which methods are most effective for instantaneous learning, successful intervention, and improved clinical outcomes.MethodsTo explore instantaneous learning in a stressful environment, we conducted a quantitative randomized controlled trial to measure how accurately individuals responded to memory-based survey questions guided by different instructional methods. Students from a large university in the Midwest (n = 157) were recruited in a public space on campus and accessed the six-question survey on their mobile devices. The intervention group competed the survey immediately while the research team created a distracting environment. The control group was asked to complete the survey later in a quiet environment.ResultsThe intervention group correctly answered 0.72 questions fewer than the control group (p = .000, CI [0.337, 1.103]). Questions Q1 and Q5 contained direct instructions with a verbal component and showed the greatest accuracy with over 90% correct for both stressful and controlled environments.ConclusionsThe variability in the responses suggests that there are environmental factors as well as instructional design features which influence instantaneous learning. The findings of this study begin to address the gap in knowledge about the effects of stress on instantaneous learning and the most effective types of instruction for untrained bystanders in emergency situations.

Development of an Educational Chatbot System for Enhancing Students’ Biology Learning Performance

&lt;p&gt;In traditional biology classrooms, teachers do not have enough class time to interact with students, and students often do not have the opportunities to raise questions in class. Therefore, out of class, students may not have sufficient ability to learn and review biology knowledge based on faulty foundations. To address the problem, it is important to provide suitable learning tools to support students conducting biology learning out of class. In recent years, chatbots have been adopted in various fields for providing instant interactions and supports to users. Therefore, by using chatbot technologies in biology education, students can interact with a chatbot out of class anytime and anywhere and further obtain instant learning supports. Bearing this in mind, the aim of this study is to propose a biology learning chatbot system that can support students to learn biology out of class. Moreover, a quasi-experiment was designed for a grade 7 biology course in a junior high school. 34 students were asked to participate in the experiment and divided into a control group and an experimental group. The experimental results indicated that the proposed chatbot system had a positive impact on students’ learning achievements in traditional biology learning.&lt;/p&gt; &lt;p&gt;&amp;nbsp;&lt;/p&gt;

PENGARUH MEDIA FILM ANIMASI NUSSA DAN RARRA TERHADAP PENDIDIKAN KARAKTER SISWA

Many moral declines occur, one of which is in the world of education, one of which is the use of smartphones in school children where they are easily influenced by trend developments through social media. So parents have to be extra in educating their children at home. In addition, the decline in ethics and morals that is happening at this time makes schools also strive to educate in order to realize the character education achieved. The government launched Strengthening Character Education (PPK) which can be integrated into the educational curriculum, one of which is through learning activities in schools using sophisticated and effective media. Researchers used the animated film Nussa and Rarra as a tool in the study. So this study aims to determine the effect of the animated film Nussa and Rarra on the character of students in PAI learning class IV SDN Karangsetia 03 Kab. Bekasi. This research is a quantitative research with the type of research design quasi experiment. With a population of all fourth grade students at SDN Karangsetia 03, the samples used were grades IV A and IV C. The data collection techniques used in this study were questionnaires and tests. The results obtained After being given treatment (treatment) and conducted a posttest. Character education of students is better using the animated film Nussa and Rarra compared to learning using power point media because it is more effective and varied, this is reinforced by the results of statistical tests assisted by SPPS version 25. The animated film media Nussa and Rarra has an effect on students' character education because the animated film Nussa and Rarra is not only funny but contains character education values ​​in each episode, then the role of the character has the potential to be imitated by children or become role models and this animated film contains a religious theme that is thick with the adab and behavior of a Muslim. The results of the attitude assessment must be understood as a process not as the result of an instant learning process that is assessed by educators every time they complete the learning process. Therefore, this assessment is an accumulative process or will continue to develop on student behavior indefinitely.

Application of Deep Reinforcement Learning to NS-SHAFT Game Signal Control.

Reinforcement learning (RL) with both exploration and exploit abilities is applied to games to demonstrate that it can surpass human performance. This paper mainly applies Deep Q-Network (DQN), which combines reinforcement learning and deep learning to the real-time action response of NS-SHAFT game with Cheat Engine as the API of game information autonomously. Based on a personal computer, we build an experimental learning environment that automatically captures the NS-SHAFT’s frame, which is provided to DQN to decide the action of moving left, moving right, or stay in same location, survey different parameters: such as the sample frequency, different reward function, and batch size, etc. The experiment found that the relevant parameter settings have a certain degree of influence on the DQN learning effect. Moreover, we use Cheat Engine as the API of NS-SHAFT game information to locate the relevant values in the NS-SHAFT game, and then read the relevant values to achieve the operation of the overall experimental platform and the calculation of Reward. Accordingly, we successfully establish an instant learning environment and instant game training for the NS-SHAFT game.

Corpus Design of Chinese Medicine English Vocabulary Translation Teaching System Based on Python

The current corpus has the problem of imperfect span retrieval function, which leads to a large classification noise. This paper designs a Python-based corpus of Chinese medicine English vocabulary translation teaching system. Here, we select the script material of web crawler, extract topic tags in the form of tag window, calculate the amount of information carried by words, use Python to extract the characteristics of Chinese medicine English vocabulary, and according to the observation value of exploration strategy, use instant time difference learning algorithm to construct the translation mode of teaching system, limit the scope of key words, and design the cross-range retrieval function of corpus. Experimental results: the average classification noise of the designed corpus and the other two corpora is 25.007[Formula: see text]dB, 33.877[Formula: see text]dB and 32.166[Formula: see text]dB, which proves that the integrated Python corpus has higher comprehensive value.

Promoting Students’ Programming Logic and Problem-Solving Awareness With Precision Feedback: A Two-Tier Test-Based Online Programming Training Approach

Fostering students' computer programming skills has become an important educational issue in the globe. However, it remains a challenge for students to understand those abstract concepts when learning computer programming, implying the need to provide instant learning diagnosis and feedback in computer programming activities. In this study, a Two-Tier Test-Based Programming Training (T3PT) approach was proposed. Accordingly, an online learning system was developed to provide students with precision feedback for guiding them to identify misconceptions of computer programming to improve their computer programming learning achievement. In order to examine the effects of the proposed approach, a learning system was developed and a quasi-experiment was conducted. Two classes of 99 eighth-grade students from Taiwan were divided into an experimental group and a control group. The students in the experimental group used the learning system based on the T3PT approach, while the control group used the conventional learning system. The experimental results showed that the proposed approach was significantly superior to the conventional programming learning approach in terms of students' programming logic concepts, problem-solving awareness, technology acceptance, and satisfaction with the learning approach. Accordingly, discussion and suggestions are provided for future research.

Pattern Inversion as a Pattern Recognition Method for Machine Learning

Artificial neural networks use a lot of coefficients that take a great deal of computing power for their adjustment, especially if deep learning networks are employed. However, there exist coefficients-free extremely fast indexing-based technologies that work, for instance, in Google search engines, in genome sequencing, etc. The paper discusses the use of indexing-based methods for pattern recognition. It is shown that for pattern recognition applications such indexing methods replace with inverse patterns the fully inverted files, which are typically employed in search engines. Not only such inversion provides automatic feature extraction, which is a distinguishing mark of deep learning, but, unlike deep learning, pattern inversion supports almost instantaneous learning, which is a consequence of absence of coefficients. The paper discusses a pattern inversion formalism that makes use on a novel pattern transform and its application for unsupervised instant learning. Examples demonstrate a view-angle independent recognition of three-dimensional objects, such as cars, against arbitrary background, prediction of remaining useful life of aircraft engines, and other applications. In conclusion, it is noted that, in neurophysiology, the function of the neocortical mini-column has been widely debated since 1957. This paper hypothesizes that, mathematically, the cortical mini-column can be described as an inverse pattern, which physically serves as a connection multiplier expanding associations of inputs with relevant pattern classes.

Instantaneous Learning in Pattern Recognition

Instantaneous Learning in Pattern Recognition

Safety lifetime analysis method for multi-mode time-dependent structural system

It is important to determine the safety lifetime of Multi-mode Time-Dependent Structural System (MTDSS). However, there is still a lack of corresponding analysis methods. Therefore, this paper establishes MTDSS safety lifetime model firstly, and then proposes a Kriging surrogate model based method to estimate safety lifetime. The first step of proposed method is to construct the Kriging model of MTDSS performance function by using extremum learning function. By identifying possible extremum mode of MTDSS, the performance function of MTDSS can be equivalently transformed into the one of Single-mode Time-Dependent Structure (STDS). The second step is to use the Advanced First Failure Instant Learning Function (AFFILF) to train the Kriging model constructed in the first step, so that the convergent Kriging model can identify the possible First Failure Instant (FFI) of STDS. Then safety lifetime can be searched quickly by dichotomy search. By using AFFILF, the minimum instant that the state is not accurately identified by the current Kriging model is selected as the training point, which avoids the unnecessary calculation which may be introduced into the existing First Failure Instant Learning Function (FFILF). In addition, the Candidate Sample Pool (CSP) reduction strategy is also adopted. By adaptively deleting the random candidate sample points whose FFI have been accurately identified by the current Kriging model, the training efficiency is further improved. Three cases show that the proposed method is accurate and efficient.

Shadow Removal Method for Single Image Based on Instant Learning

Shadow Removal Method for Single Image Based on Instant Learning

Awesome back-propagation machine learning paradigm

For a better future in machine learning (ML), it is necessary to modify our current concepts to get the fastest ML. Many designers had attempted to find the optimal learning rates in their applications through many algorithms over the past decades, but they have not yet achieved their target of highest speed of back-propagation (BP). This research proposes a novel BP rule called the Instant Learning Ratios-Machine Learning (ILR-ML) or (ILRML). Unlike the traditional BP algorithms, the ILR-ML offers its learning without the concepts of the learning rate(s). The ILR-ML has a new concept called the "Learning Ratio" and indicated by a sign (Δl). The ILR-ML performs the full BP algorithm with 100% accuracy per each learning iteration. The ILR-ML is more suitable for the online machine learning.

Effectiveness of Transactional Communication in the Implementation of Collegiate Curriculum

The interaction between lecturers and students will be effective if delivered through transactional communication so that they can mutually encode and decode what is communicated about learning. In today’s development of the era of digital communication technology, students tend to get instant learning resources by using their gadgets instead of communicating with their lecturers. The implementation of collegiate curriculum refers to the Indonesian national qualification framework (KKNI) which will be effective with the student centered learning (SCL) approach. Lecturers act as active facilitators and students are active, mutually meaningful communication occurs so that the application of collegiate curriculum can be effectively achieved. The purpose of this study is to have description about how transactional communication is implemented in achieving learning effectiveness. The research method used is descriptive qualitative with a case study research approach. The research paradigm is constructivism. The theories used are cognitive theory and behavioral theory. The results of the research show that the implementation of the Collegiate Curriculum, namely planning, implementation, learning and learning outcomes are effective by using transactional communication

Adaptive DTN Routing: A Neuromorphic Networking Perspective

Routing is one of the main drivers of the end-to-end performance of bundle transmissions over a disruption tolerant network given the potentially large impact of the temporary but long-term partitioning that can occur at different sections of the network. A neuromorphic networking approach that defines an adaptive bundle routing for disruption-tolerant networks (DTN) is proposed where spiking neuronal networks (SNN) are used to determine the routing decisions of autonomous agents. The event-driven information encoding of spiking neurons involves very low energy consumption, which makes this approach attractive for challenging DTN applications with limited access to energy sources. The SNNs are continually updated within an autonomic loop, which produces synapse strength updates that are proportional to the expected communication costs of the routing decisions. A reward shaping procedure and a delay-tolerant mechanism for finding the local link-state is proposed, which allows determining instantaneous learning rewards for the agents. The method was tested on an emulated space communications network with scheduled disruptions. The results show that the proposed cognitive routing approach offers improved bundle delivery performance under network congestion compared to the standard Contact Graph Routing.

Sequential online prediction in the presence of outliers and change points: An instant temporal structure learning approach

In this paper, we consider sequential online prediction (SOP) for streaming data in the presence of outliers and change points. We propose an INstant TEmporal structure Learning (INTEL) algorithm to address this problem. Our INTEL algorithm is developed based on a full consideration of the duality between online prediction and anomaly detection. We first employ a mixture of weighted Gaussian process models (WGPs) to cover the expected possible temporal structures of the data. Then, based on the rich modeling capacity of this WGP mixture, we develop an efficient technique to instantly learn (capture) the temporal structure of the data that follows a regime shift. This instant learning is achieved only by adjusting one hyper-parameter of the mixture model. A weighted generalization of the product of experts (POE) model is used for fusing predictions yielded from multiple GP models. An outlier is declared once a real observation seriously deviates from the fused prediction. If a certain number of outliers are consecutively declared, then a change point is declared. Extensive experiments are performed using a diverse of real datasets. Results show that the proposed algorithm is significantly better than benchmark methods for SOP in the presence of outliers and change points.

Investigating the use of mobile instant messaging-facilitated 5E-flipped learning: a two-stage study

Despite the worldwide popularity of flipped learning, we understand little about how to coherently structure a flipped class with appropriate communication services to engage students. This two-stage study examines in flipped learning context, how students perceived learning guided by the 5E design model (engagement-exploration-explanation-elaboration-evaluation), and how different technologies impacted the learning experience. We first reported a multiple-case study on how students felt about the 5E model-based flipped learning, facilitated by mobile instant messaging (MIM) tools. We then conducted a comparative study to investigate how students felt about using either MIM or a threaded asynchronous forum for discussion. The results suggested that 92% of participants found flipped learning more engaging and 81% reported that flipped learning gave them more time to discuss or solve problems. Most students welcomed using MIM for flipped learning, but online forum was still preferred for academic discussions. Possible reasons and suggestions for future research are discussed.