Human Learning Behavior Research Articles

Pain reflects the informational value of nociceptive inputs.

Pain perception and its modulation are fundamental to human learning and adaptive behavior. This study investigated the hypothesis that pain perception is tied to pain's learning function. Thirty-one participants performed a threat conditioning task where certain cues were associated with a possibility of receiving a painful electric shock. The cues that signaled potential pain or safety were regularly changed, requiring participants to continually establish new associations. Using computational models, we quantified participants' pain expectations and prediction errors throughout the task and assessed their relationship with pain perception and electrophysiological responses. Our findings suggest that subjective pain perception increases with prediction error, that is, when pain was unexpected. Prediction errors were also related to physiological nociceptive responses, including the amplitude of nociceptive flexion reflex and electroencephalography markers of cortical nociceptive processing (N1-P2-evoked potential and gamma-band power). In addition, higher pain expectations were related to increased late event-related potential responses and alpha/beta decreases in amplitude during cue presentation. These results further strengthen the idea of a crucial link between pain and learning and suggest that understanding the influence of learning mechanisms in pain modulation could help us understand when and why pain perception is modulated in health and disease.

A deeper knowledge tracking model integrating cognitive theory and learning behavior

Knowledge tracing (KT), which aims to trace human knowledge learning process by using machines, has widely applied in online learning systems. It dynamically models student’s knowledge states in relation to different learning factors through their learning interactions. Recently, KT has attracted many researches attention due to its good performance to using deep learning. Although most of KT models have shown outstanding results, they have limitations: either ignore the human cognitive law and learning behavior, or lack the ability to go deeper modeling to trace knowledge state. In this paper, we propose a deeper knowledge tracking model integrating cognitive theory and learning behavior (CLDKT). It united the advantages of memory network and recurrent neural network of the existing deep learning KT models for modeling student learning. To better implement CLDKT, we add the residual network (ResNet) to realize the deep modeling of learning behaviors. Extensive experiments on three open benchmark datasets to evaluate our model. Experimental results demonstrate that (I) CLDKT outperforms the state-of-the-art KT models on students’ performance prediction. (II) CLDKT can deeper modeling to trace knowledge state owing to the ResNet import. (III) CLDKT has better interpretability and predictability, which proves the effectiveness of the knowledge tracing model integrating cognitive law and learning behavior.

Research on feature coding theory and typical application analysis in machine learning algorithms

Nowadays, the world is still in the environment of economic depression. In order to promote economic recovery, improve Relations of production and production efficiency, stimulate consumption expansion and upgrading, and accelerate industrial transformation and upgrading, problems such as industrial upgrading need to be solved urgently. Solving the above problems requires more useful tools, and artificial intelligence is one of them. Machine learning is the key to distinguishing artificial intelligence from ordinary program code. Unlike people learning knowledge, machine learning has its own unique language algorithms and behavioral logic. Machine learning, as a technology active in the field of artificial intelligence in recent years, specializes in studying how computers learn, simulate and realize part of human learning behavior, so as to provide data mining and behavior prediction for humans, to obtain new knowledge or skills, or to strengthen the original basic ability of machines. In this study, a variety of common coding algorithms and learning strategies in machine learning are discussed, supervised learning algorithms are selected as examples in the learning strategies, models are further selected and evaluated for a variety of algorithms, and parameters are adjusted and performance is analyzed. As for the theoretical analysis in the research, the paper makes a tentative application in the three fields of housing price, physical store sales and digital recognition, explores and selects the corresponding application method in the appropriate scenario, and expands the application field of machine learning.

How well do rudimentary plasticity rules predict adult visual object learning?

A core problem in visual object learning is using a finite number of images of a new object to accurately identify that object in future, novel images. One longstanding, conceptual hypothesis asserts that this core problem is solved by adult brains through two connected mechanisms: 1) the re-representation of incoming retinal images as points in a fixed, multidimensional neural space, and 2) the optimization of linear decision boundaries in that space, via simple plasticity rules applied to a single downstream layer. Though this scheme is biologically plausible, the extent to which it explains learning behavior in humans has been unclear-in part because of a historical lack of image-computable models of the putative neural space, and in part because of a lack of measurements of human learning behaviors in difficult, naturalistic settings. Here, we addressed these gaps by 1) drawing from contemporary, image-computable models of the primate ventral visual stream to create a large set of testable learning models (n = 2,408 models), and 2) using online psychophysics to measure human learning trajectories over a varied set of tasks involving novel 3D objects (n = 371,000 trials), which we then used to develop (and publicly release) empirical benchmarks for comparing learning models to humans. We evaluated each learning model on these benchmarks, and found those based on deep, high-level representations from neural networks were surprisingly aligned with human behavior. While no tested model explained the entirety of replicable human behavior, these results establish that rudimentary plasticity rules, when combined with appropriate visual representations, have high explanatory power in predicting human behavior with respect to this core object learning problem.

Superlow Power Consumption Memristor Based on Borphyrin-Deoxyribonucleic Acid Composite Films as Artificial Synapse for Neuromorphic Computing.

Memristor synapses based on green and pollution-free organic materials are expected to facilitate biorealistic neuromorphic computing and to be an important step toward the next generation of green electronics. Metalloporphyrin is an organic compound that widely exists in nature with good biocompatibility and stable chemical properties, and has already been used to fabricate memristors. However, the application of metalloporphyrin-based memristors as synaptic devices still faces challenges, such as realizing a high switching ratio, low power consumption, and bidirectional conductance modulation. We developed a memristor that improves the resistive switching (RS) characteristics of Zn(II)meso-tetra(4-carboxyphenyl) porphine (ZnTCPP) by combining it with deoxyribonucleic acid (DNA) in a composite film. The as-fabricated ZnTCPP-DNA-based device showed excellent RS memory characteristics with a sufficiently high switching ratio of up to ∼104, super low power consumption of ∼39.56 nW, good cycling stability, and data retention capability. Moreover, bidirectional conductance modulation of the ZnTCPP-DNA-based device can be controlled by modulating the amplitudes, durations, and intervals of positive and negative pulses. The ZnTCPP-DNA-based device was used to successfully simulate a series of synaptic functions including long-term potentiation, long-term depression, spike time-dependent plasticity, paired-pulse facilitation, excitatory postsynaptic current, and human learning behavior, which demonstrates its potential applicability to neuromorphic devices. A two-layer artificial neural network was used to demonstrate the digit recognition ability of the ZnTCPP-DNA-based device, which reached 97.22% after 100 training iterations. These results create a new avenue for the research and development of green electronics and have major implications for green low-power neuromorphic computing in the future.

Integrating Curricula with Replays: Its Effects on Continual Learning

Humans engage in learning and reviewing processes with curricula when acquiring new skills or knowledge. This human learning behavior has inspired the integration of curricula with replay methods in continual learning agents. The goal is to emulate the human learning process, thereby improving knowledge retention and facilitating learning transfer. Existing replay methods in continual learning agents involve the random selection and ordering of data from previous tasks, which has shown to be effective. However, limited research has explored the integration of different curricula with replay methods to enhance continual learning. Our study takes initial steps in examining the impact of integrating curricula with replay methods on continual learning in three specific aspects: the interleaved frequency of replayed exemplars with training data, the sequence in which exemplars are replayed, and the strategy for selecting exemplars into the replay buffer. These aspects of curricula design align with cognitive psychology principles and leverage the benefits of interleaved practice during replays, easy-to-hard rehearsal, and exemplar selection strategy involving exemplars from a uniform distribution of difficulties. Based on our results, these three curricula effectively mitigated catastrophic forgetting and enhanced positive knowledge transfer, demonstrating the potential of curricula in advancing continual learning methodologies. Our code and data are available: https://github.com/ZhangLab-DeepNeuroCogLab/Integrating-Curricula-with-Replays

Machine learning and lumbar spondylolisthesis

While lumbar spondylolisthesis is one of the most common conditions cared for by spine surgeons, there remains limited evidence guiding its diagnosis, classification, and management. The diversity in its underlying causes, clinical manifestations, and spinal anatomical variations poses a particular challenge in making informed clinical decisions. Machine learning (ML) methods offer novel opportunities to address these challenges by leveraging data-driven approaches. This chapter provides a comprehensive overview of the potential applications of ML in the field of lumbar spondylolisthesis. ML is a branch of artificial intelligence that employs statistical algorithms to mimic human learning behavior. In the domain of diagnosis, ML methods have been applied to detect spondylolisthesis using medical imaging. In particular, deep learning models have shown high accuracy in detecting spondylolisthesis from X-rays and MRIs, suggesting ML's potential as a diagnostic tool. Additionally, ML can aid in distinguishing spondylolisthesis grades and subtypes. Although automatic grading remains challenging, recent advances suggest that emerging ML techniques may be effective in classifying spondylolisthesis subtypes and guiding subsequent decision-making. Already, ML has been used to predict spondylolisthesis treatment outcomes, such as functional recovery and hospital length of stay. While promising, most of these prediction studies used "shallow" ML techniques, suggesting that further gains may be realized by applying deep learning methods to larger, complex datasets. In conclusion, ML advances hold promise in spondylolisthesis diagnosis, classification, and outcome prediction. In the future, these methods may help support more personalized and effective management of lumbar spondylolisthesis.

A Multigoal Path-Planning Approach for Explosive Ordnance Disposal Robots Based on Bidirectional Dynamic Weighted-A* and Learn Memory-Swap Sequence PSO Algorithm

In order to protect people’s lives and property, increasing numbers of explosive disposal robots have been developed. It is necessary for an explosive ordinance disposal (EOD) robot to quickly detect all explosives, especially when the location of the explosives is unknown. To achieve this goal, we propose a bidirectional dynamic weighted-A star (BD-A*) algorithm and a learn memory-swap sequence particle swarm optimization (LM-SSPSO) algorithm. Firstly, in the BD-A* algorithm, our aim is to obtain the shortest distance path between any two goal positions, considering computation time optimization. We optimize the computation time by introducing a bidirectional search and a dynamic OpenList cost weight strategy. Secondly, the search-adjacent nodes are extended to obtain a shorter path. Thirdly, by using the LM-SSPSO algorithm, we aim to plan the shortest distance path that traverses all goal positions. The problem is similar to the symmetric traveling salesman problem (TSP). We introduce the swap sequence strategy into the traditional PSO and optimize the whole PSO process by imitating human learning and memory behaviors. Fourthly, to verify the performance of the proposed algorithm, we begin by comparing the improved A* with traditional A* over different resolutions, weight coefficients, and nodes. The hybrid PSO algorithm is also compared with other intelligent algorithms. Finally, different environment maps are also discussed to further verify the performance of the algorithm. The simulation results demonstrate that our improved A* algorithm has superior performance by finding the shortest distance with less computational time. In the simulation results for LM-SSPSO, the convergence rate significantly improves, and the improved algorithm is more likely to obtain the optimal path.

Spatial Differentiation Characteristics of Rural Areas Based on Machine Learning and GIS Statistical Analysis—A Case Study of Yongtai County, Fuzhou City

With the development of machine learning and GIS (geographic information systems) technology, it is possible to combine them to mine the knowledge rules behind massive spatial data. GIS, also known as geographic information systems, is a comprehensive discipline, which combines geography and cartography and has been widely used in different fields. It is a computer system for inputting, storing, querying, analyzing, and displaying geographic data. This paper mainly studies the spatial differentiation characteristics of rural areas based on machine learning (ML) and GIS statistical analysis. This paper studies 21 township units in Yongtai County. In this paper, ENVI remote sensing image processing software is used to carry out the geometric correction of Landsat-8 remote sensing data. ML is multidisciplinary and interdisciplinary, involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, and other disciplines. It is specialized in studying how computers simulate or realize human learning behavior to obtain new knowledge or skills, and reorganize existing knowledge structures to continuously improve its own performance. The purpose of using band fusion is to provide more data information for the study and improve the accuracy of land classification results. Through the extraction of evaluation elements, this paper preliminarily confirms the evaluation index object of a rural human settlement environment evaluation system from the perspective of spatial layout rationality. This paper uses a VMD-GWO-ELM-based three-stage evolutionary extreme learning machine evaluation method to simulate the model. In the same way, when the model is trained again, extra weight is given to extract the feature points to reduce the similarity. Experimental data show that GWO-SVM has good classification performance, with the cross-validation rate reaching 91.66% and the recognition rate of test samples reaching 82.41%. The results show that GIS statistics can provide a reference for environmental protection, which is conducive to land-use planning, implementation of environmental impact assessment of land-use planning, and ultimately achieving sustainable development.

Special Physical Characteristic and Training Strategy of Badminton Based on Machine Learning Algorithm

Machine learning algorithms involve many fields such as statistics and probability theory. It is mainly aimed at computer simulation or execution of human learning behaviors to acquire new knowledge and skills. And the existing knowledge is to be integrated and then continue to be improved. This study aims to introduce a machine learning algorithm to establish a corresponding training effect evaluation model, to use the model to research and analyze the special physical fitness training strategy of badminton, and to discuss the appropriate badminton physical fitness training method. At the same time, the Markov model is used to compare the effect of the machine learning algorithm model. The experimental results show that the combination of four kinds of training of phosphate energy supply capacity, speed, sensitivity, and core strength can improve the training performance of badminton players by 10 points and the training efficiency by 4%. Therefore, it is necessary to comprehensively and reasonably arrange the four physical training methods to achieve better training effects. The average evaluation and prediction accuracies of the machine learning algorithm model were 86% and 87%, respectively, and the computing time was 26.7 ms and 13.6 ms. The average evaluation and prediction accuracies of the Markov model were 79% and 82%, respectively, and the computing time was 19.5 ms and 11.8 ms. It can be seen that the model based on the machine learning algorithm is more accurate than the Markov model.

Development Countermeasures of Business Management Informatization Based on Machine Learning Algorithm

Business management is an important part of social service and governance. With the development of economy and technology, the pressure on enterprises to participate in the market competition is gradually increasing. Therefore, it is particularly important to establish an information-based business management system. However, there are still many problems in the process of informatization development of enterprise business management, which seriously restricts the comprehensive development of enterprises. On the one hand, the existing business management information system lacks effective security and supervision links. On the other hand, there is a lack of professional information technology personnel within the enterprise. Machine learning is a discipline that specializes in how to simulate human learning behavior, and it can continuously learn new knowledge structures to make itself progress and develop. The development of enterprise business management informatization is consistent with the development direction of machine learning itself. Therefore, this paper proposes a business management informatization construction method based on a machine learning algorithm, which aims to further promote the development of enterprises in the direction of informatization and modernization and improve the informatization level of enterprises. Experiments show that with the support of machine learning technology, the level of enterprise business management informatization can reach 82.31%, fully indicating that machine learning can help enterprises innovate management models and improve the level of information management of enterprises.

Application of Machine Learning Algorithms in the Development and Consumption Trend of Green and Intelligent Vehicles under the Background of Big Data

In the face of global warming, air pollution, and other difficulties, electric vehicles have become an industry strongly supported by various countries due to their good environmental protection characteristics. In the context of big data, people are exposed to more and more information, and the convenience brought by big data is also increasing. Based on this background, the development of green and intelligent vehicles is getting faster and faster. This paper is aimed at studying the application of machine learning algorithms in the development and consumption trends of green and intelligent vehicles in the context of big data. This paper proposes machine learning algorithms based on big data, as well as support vector machine algorithms and so on. Machine learning algorithms specialize in how computers simulate or implement human learning behaviors to acquire new knowledge or skills and to reorganize existing knowledge structures to continuously improve their performance. The test results of this paper show that, starting from 2014, China has begun to vigorously develop green and intelligent vehicles. In 2014, the production volume of green and intelligent vehicles in China was 3,675, and the sales volume was 2,790. The development of green and intelligent vehicles is not very good and has not been fully accepted by the public. However, since 2017, the production and sales of green and intelligent vehicles have been slowly increasing. By 2020, the production of green and intelligent vehicles will be 24,360 and the sales will be 24,090. It can be seen that with the development of time, green and intelligent vehicles are gradually being recognized.

Reward prediction errors, not sensory prediction errors, play a major role in model selection in human reinforcement learning

Model-based reinforcement learning enables an agent to learn in variable environments and tasks by optimizing its actions based on the predicted states and outcomes. This mechanism has also been considered in the brain. However, exactly how the brain selects an appropriate model for confronting environments has remained unclear. Here, we investigated the model selection algorithm in the human brain during a reinforcement learning task. One primary theory of model selection in the brain is based on sensory prediction errors. Here, we compared this theory with an alternative possibility of internal model selection with reward prediction errors. To compare these two theories, we devised a switching experiment from a first-order Markov decision process to a second-order Markov decision process that provides either reward- or sensory prediction error regarding environmental change. We tested two representative computational models driven by different prediction errors. One is the sensory prediction-error-driven Bayesian algorithm, which has been discussed as a representative internal model selection algorithm in the animal reinforcement learning task. The other is the reward-prediction-error-driven policy gradient algorithm. We compared the simulation results of these two computational models with human reinforcement learning behaviors. The model fitting result supports that the policy gradient algorithm is preferable to the Bayesian algorithm. This suggests that the human brain employs the reward prediction error to select an appropriate internal model in the reinforcement learning task.

3D Reconstruction Method of Virtual and Real Fusion Based on Machine Learning

With the continuous development of computer vision technology, people are paying more and more attention to the method of using computers to simulate actual 3D scenes, and the requirements for 3D reconstruction technology are getting higher and higher. Virtual and real fusion refers to combining the virtual environment generated by the computer with the actual scenes around the user through photoelectric display, sensors, computer graphics, multimedia, and other technologies. This is a technology that can obtain more convenient and direct expressions, and it is also a technique for expressing content more abundantly and accurately. The key to virtual and real fusion technology is the registration of virtual objects and real scenes. It means that the system should be able to correctly estimate the position and posture of the camera in the real world, and then place the virtual object where it should be. Machine learning is a multifield interdisciplinary subject that specializes in how computers simulate or realize human learning behaviors. It is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are in all the fields of artificial intelligence. This article introduces the virtual-real fusion 3D reconstruction method based on machine learning, compares the performance of the method with other algorithms through experiments, and draws the following conclusion: the algorithm in this study is the fastest, with an average speed of 72.9% under different times. To evaluate the image acquisition indicators of each algorithm, the algorithm in this study has the lowest error rate. The matching accuracy of each algorithm is tested, and it is found that the average matching accuracy of the algorithm in this study is about 0.87, which is the highest.

Appositeness of Optimized and Reliable Machine Learning for Healthcare: A Survey.

Machine Learning (ML) has been categorized as a branch of Artificial Intelligence (AI) under the Computer Science domain wherein programmable machines imitate human learning behavior with the help of statistical methods and data. The Healthcare industry is one of the largest and busiest sectors in the world, functioning with an extensive amount of manual moderation at every stage. Most of the clinical documents concerning patient care are hand-written by experts, selective reports are machine-generated. This process elevates the chances of misdiagnosis thereby, imposing a risk to a patient's life. Recent technological adoptions for automating manual operations have witnessed extensive use of ML in its applications. The paper surveys the applicability of ML approaches in automating medical systems. The paper discusses most of the optimized statistical ML frameworks that encourage better service delivery in clinical aspects. The universal adoption of various Deep Learning (DL) and ML techniques as the underlying systems for a variety of wellness applications, is delineated by challenges and elevated by myriads of security. This work tries to recognize a variety of vulnerabilities occurring in medical procurement, admitting the concerns over its predictive performance from a privacy point of view. Finally providing possible risk delimiting facts and directions for active challenges in the future.

More Negative FRN From Stopping Searches Too Late Than Too Early: An ERP Study.

It is widely known that the feedback from a decision outcome may evoke emotions like regret, which results from a comparison between the gain the decision-maker has made and the gain he/she might make. Less is known about how search behavior is linked to feedback in a sequential search task such as searching for jobs, employees, prices, investments, disinvestments, or other items. What are the neural responses once subjects decide to stop searching and receive the feedback that they stopped too early or too late compared with the optimal stopping time? In an experimental setting of a search task, we found that the feedback-related negativity (FRN) induced by the feedback from stopping too late was more negative than stopping too early, suggesting that subjects might experience stronger regret when stopping too late. Subjects preferred to stop searching earlier if the last feedback was that they stopped too late, and vice versa, although they did not always benefit more from such adjustment. This might reflect general patterns of human learning behavior, which also manifests in many other decisions. Gender differences and risk attitudes were also considered in the study.

Modelling software reliability growth through generalized inflection S-shaped fault reduction factor and optimal release time

The fault reduction factor (FRF) is a significant parameter for controlling the software reliability growth. It is the ratio of net fault correction to the number of failures encountered. In literature, many factors affect the behaviour of FRF, namely fault dependency, debugging time-lag, human learning behaviour and imperfect debugging. Besides this, several distributions, for example, inflection S-shaped, Weibull and Exponentiated-Weibull, are used as FRF. However, these standard distributions are not flexible to describe the observed behaviour of FRFs. This paper proposes three different software reliability growth models (SRGMs), which incorporate a three-parameter generalized inflection S-shaped (GISS) distribution as FRF. To model realistic SRGMs, time lags between fault detection and fault correction processes are also incorporated. This study proposed two models for the single release, whereas the third model is designed for multi-release software. Moreover, the first model is in perfect debugging, while the rest of the two are in an imperfect debugging environment. The extensive experiments are conducted for the proposed models with six single release and one multi-release data-sets. The choice of GISS distribution as an FRF improves the software reliability evaluation in comparison with the existing systems in the literature. Finally, the development cost and optimal release time are calculated in a perfect debugging environment.

Mode Optimization and Rule Management of Intellectual Property Rights Protection of Educational Resource Data Based on Machine Learning Algorithm

Educational resource data are a collection of final documents obtained by users, including full-text journals, books, dissertations, newspapers, conference papers, and other database materials. While searching for information in the educational resource database, these resources also have functions such as copying, downloading, reproduction, and dissemination, which raise the issue of expression and protection of intellectual property. Machine learning takes how computers simulate human learning behaviors as the main research content, which can independently determine learning objects, construct their characteristics, perform additional operations beyond the limitations of preset instructions, and discover value from the expression of relative works. On the basis of summarizing and analyzing previous research works, this paper expounded the current research status and significance of intellectual property expression and protection of educational resource data; elaborated the development background, current status, and future challenges of machine learning technology; introduced the methods and principles of data classification algorithm and protection authority identification; performed the technical framework design and expression system establishment of the intellectual property expression of educational resource data based on machine learning; analyzed the mode optimization and rule management of intellectual property protection of educational resource data based on machine learning; and finally conducted a simulation experiment and its result analysis. The results show that the machine learning technology can build a subject-oriented, highly integrated, and time-changing educational resource data storage environment; the comprehensive, analysis-oriented decision-supporting system formed by machine learning can give full play to the potential role of data integration and value discovery and is therefore of great significance for the intellectual property expression and protection of integrated and complexly-related educational resource data. The study results of this paper provide a reference for further research on the intellectual property expression and protection of educational resource data based on machine learning.

Progressive Mimic Learning: A new perspective to train lightweight CNN models

Knowledge distillation (KD) builds a lightweight Student Model (SM) and trains it to approximate a large Teacher Model (TM) by exploring knowledge learned by the TM, which shows effectiveness to train lightweight CNN models. However, training a small SM to achieve better performance remains a challenging problem. Recent researches on human learning behaviors show that both the knowledge from teachers and the knowledge learning processes of teachers are significant for students. Inspired by this characteristic, in this paper, we propose a new perspective, called Progressive Mimic Learning (PML), to train lightweight CNN models by mimicking the learning trajectory of the TM. In order to obtain a more powerful SM, the useful hints in the learning process of the TM are explored. To start with, the TM learning process is divided into multiple stages, and the last state of the TM in each stage is recorded as a landmark. The learning trajectory of the TM is composed of these landmarks. Then, a landmark loss is defined to constrain the SM to progressively mimic the learning process of the TM, by employing landmarks in the learning trajectory as a training hint of the SM. Several experiments are conducted on four benchmark data sets, CIFAR-10, CIFAR-100, Fashion-MNIST, and ImageNet-10, to investigate the performance of the PML. The results show that the PML can make SMs generate more accurate predictions than SMs trained by its counterparts.

Hybrid of human learning optimization algorithm and particle swarm optimization algorithm with scheduling strategies for the flexible job-shop scheduling problem

The flexible job-shop scheduling problem (FJSP) is a well-known combinational optimization problem. Studying FJSP is essential for promoting production efficiency and effectiveness. Different kinds of improved particle swarm optimization (PSO) algorithms have produced superior results for FJSP in the last few decades. Meanwhile, the human learning optimization (HLO) algorithm, a simple and adaptive learning algorithm for learning system, has helped improve algorithm performance by imitating human learning behavior in recent research. The study proposes a hybrid HLO-PSO algorithm, which utilizes various combinations of the proposed improved PSO and proposed scheduling strategies to solve FJSP under the algorithm architecture of HLO. With the guidance of HLO, the individual learning ability of every particle is further promoted based on the existed advantage of collective action decision of PSO; and with the help of rule-based scheduling strategies, the search capacity of the proposed improved PSO is also further enhanced. By the detailed exposition and analysis, the proposed HLO-PSO is easily implemented and embedded in other production system software or learning system software. Meanwhile, by using it to solve several groups of FJSP instances, the result comparisons with other related algorithms reveal that HLO-PSO can efficiently solve most of single-objective FJSP.