Learning Problem‐Solving Rules as Search Through a Hypothesis Space

Learning to solve a class of problems can be characterized as a search through a space of hypotheses about the rules for solving these problems. A series of four experiments studied how different learning conditions affected the search among hypotheses about the solution rule for a simple computational problem. Experiment 1 showed that a problem property such as computational difficulty of the rules biased the search process and so affected learning. Experiment 2 examined the impact of examples as instructional tools and found that their effectiveness was determined by whether they uniquely pointed to the correct rule. Experiment 3 compared verbal directions with examples and found that both could guide search. The final experiment tried to improve learning by using more explicit verbal directions or by adding scaffolding to the example. While both manipulations improved learning, learning still took the form of a search through a hypothesis space of possible rules. We describe a model that embodies two assumptions: (1) the instruction can bias the rules participants hypothesize rather than directly be encoded into a rule; (2) participants do not have memory for past wrong hypotheses and are likely to retry them. These assumptions are realized in a Markov model that fits all the data by estimating two sets of probabilities. First, the learning condition induced one set of Start probabilities of trying various rules. Second, should this first hypothesis prove wrong, the learning condition induced a second set of Choice probabilities of considering various rules. These findings broaden our understanding of effective instruction and provide implications for instructional design.     

Aspiration-Based and Reciprocity-Based Rules in Learning Dynamics for Symmetric Normal-Form Games

Psychologically based rules are important in human behavior and have the potential of explaining equilibrium selection and separatrix crossings to a payoff dominant equilibrium in coordination games. We show how a rule learning theory can easily accommodate behavioral rules such as aspiration-based experimentation and reciprocity-based cooperation and how to test for the significance of additional rules. We confront this enhanced rule learning model with experimental data on games with multiple equilibria and separatrix-crossing behavior. Maximum likelihood results do not support aspiration-based experimentation or anticipated reciprocity as significant explanatory factors, but do support a small propensity for non-aspiration-based experimentation by random belief and non-reciprocity-based cooperation.     

如何利用开源工具分析fMRI数据：基于多体素模式的有监督机器学习算法介绍

摘要：目前,多体素模式分析（MVPA）日渐普遍地应用于脑影像研究。近些年,机器学习的模式分类等算法在MVPA方法中被广泛应用,因其具有能够抽取高维数据模式,提高数据利用率的优点。其中一种典型的应用是利用解码的思想来解决神经表征问题,本文主要介绍了利用基于Python语言的工具库中有监督学习算法分析数据的过程。除介绍Nilearn结合Scikit-learn分析数据的步骤外,还比较不同算法的效率,为算法的选择及参数设备提供具体参考。     

A neuro-fuzzy network to generate human-understandable knowledge from data

Neuro-fuzzy networks have been successfully applied to extract knowledge from data in the form of fuzzy rules. However, one drawback with the neuro-fuzzy approach is that the fuzzy rules induced by the learning process are not necessarily understandable. The lack of readability is essentially due to the high dimensionality of the parameter space that leads to excessive flexibility in the modification of parameters during learning. In this paper, to obtain readable knowledge from data, we propose a new neuro-fuzzy model and its learning algorithm that works in a parameter space with reduced dimensionality. The dimensionality of the new parameter space is necessary and sufficient to generate human-understandable fuzzy rules, in the sense formally defined by a set of properties. The learning procedure is based on a gradient descent technique and the proposed model is general enough to be applied to other neuro-fuzzy architectures. Simulation studies on a benchmark and a real-life problem are carried out to embody the idea of the paper.     

Personality Research and Assessment in the Era of Machine Learning

The increasing availability of high-dimensional, fine-grained data about human behaviour, gathered from mobile sensing studies and in the form of digital footprints, is poised to drastically alter the way personality psychologists perform research and undertake personality assessment. These new kinds and quantities of data raise important questions about how to analyse the data and interpret the results appropriately. Machine learning models are well suited to these kinds of data, allowing researchers to model highly complex relationships and to evaluate the generalizability and robustness of their results using resampling methods. The correct usage of machine learning models requires specialized methodological training that considers issues specific to this type of modelling. Here, we first provide a brief overview of past studies using machine learning in personality psychology. Second, we illustrate the main challenges that researchers face when building, interpreting, and validating machine learning models. Third, we discuss the evaluation of personality scales, derived using machine learning methods. Fourth, we highlight some key issues that arise from the use of latent variables in the modelling process. We conclude with an outlook on the future role of machine learning models in personality research and assessment.     

Perspective on “in the wild” movement analysis using machine learning

Recent advances in wearable sensing and machine learning have created ample opportunities for “in the wild” movement analysis in sports, since the combination of both enables real-time feedback to be provided to athletes and coaches, as well as long-term monitoring of movements. The potential for real-time feedback is useful for performance enhancement or technique analysis, and can be achieved by training efficient models and implementing them on dedicated hardware. Long-term monitoring of movement can be used for injury prevention, among others. Such applications are often enabled by training a machine learned model from large datasets that have been collected using wearable sensors. Therefore, in this perspective paper, we provide an overview of approaches for studies that aim to analyze sports movement “in the wild” using wearable sensors and machine learning. First, we discuss how a measurement protocol can be set up by answering six questions. Then, we discuss the benefits and pitfalls and provide recommendations for effective training of machine learning models from movement data, focusing on data pre-processing, feature calculation, and model selection and tuning. Finally, we highlight two application domains where “in the wild” data recording was combined with machine learning for injury prevention and technique analysis, respectively.     

Accounting for Graded Performance within a Discrete Search Framework

This article presents a process account of some typicality effects and related similarity-dependent accuracy and response time phenomena that arise in the context of supervised concept acquisition. We describe Symbolic Concept Acquisition (SCA), a computational system that acquires and activates category prediction rules. In contrast to gradient representations, SCA performs by probing for prediction rules in a series of discrete steps. For learning new rules, it acquires general rules but then incrementally learns more specific ones. In describing SCA, we emphasize its functionality in terms of accuracy and efficiency and motivate its design within the set of symbolic mechanisms and memory structures defined by the Soar architecture (Laird, Newell & Rosenbloom, 1987). For replicating human behavior, we first show how SCA exhibits some typicality effects in the course of learning responding faster and more accurately to more typical test examples. Then, using data from human experiments, we evaluate SCA's qualitative predictions on accuracy and response time on individual dataset instances. We show how SCA's predictions correlate with human data across three experimental conditions concerning the effect of instruction on learning strategy.     

机器学习在儿童创伤后应激障碍识别及转归预测中的应用

创伤后应激障碍(PTSD)会给儿童发展带来负面效应,其影响甚至延续至成年期。然而传统诊断方式难以做到快速、客观、准确的识别和诊断儿童PTSD,机器学习作为一种处理大量变量和数据的新兴方法,逐渐被应用到儿童PTSD的早期预测、识别及辅助诊断等研究中。机器学习凭借其性能、原理等方面的优势,可被应用在儿童PTSD的识别与转归领域。相比自我报告式的诊断,通过机器学习辅助识别和诊断儿童PTSD的过程具有效率高、客观准确、节约资源等独特优势。然而,机器学习也在硬件成本、算法选择和预测准确度等方面存在局限性。未来研究人员需要进一步提高机器学习诊断识别儿童PTSD的准确率,并将机器学习算法同传统诊断方法结合进行更多的探索和应用。     

基于社交媒体数据的心理指标识别建模:机器学习的方法

心理指标识别建模是基于海量数据结合计算机机器学习算法识别心理特征的一种新兴方式。由于传统纸笔测量方式所存在的诸多限制,本文对基于社交媒体数据的心理建模方法及应用于心理测量的可行性进行综述,介绍了特征及提取方法、常用机器学习算法以及应用场景,并对心理指标识别建模的优势和不足进行了总结与展望。该测量方法基于社交媒体数据,相比自我报告法具有时效性高、可回溯测量、生态效度好等独特优势。然而,基于社交媒体的心理指标识别建模方法也在学习成本、硬件成本等方面存在局限性。未来研究人员需要进一步探索社会媒体信息与用户心理变量间的关联机制,并将心理指标识别模型同传统心理学研究方法结合进行更多的探索和应用。心理指标识别建模结合心理测量基本原理和计算机领域机器学习的技术,将为心理学研究打开一扇新的大门。     

The Neuropsychological Bases of Category Learning

There is a long history of research into how people learn new categories of objects or events. Recent advances have led to new insights about the neuropsychological basis of this critically important cognitive process. In particular, there is now good evidence that the frontal cortex and basal ganglia contribute to category learning, that medial temporal lobe structures make a more minor contribution, and that categorization rules are not represented in visual cortex. There is also strong evidence that normal category learning is mediated by at least two separate systems. A recent neuropsychological theory of category learning that is consistent with these data is described.     

Do consumers ever learn? Analysis of segment behavior in experimental markets

Economists argue that, despite cognitive limitations, economic agents arrive at optimal choice rules by learning. The assumption is that consumers, for example, are adaptively rational. Adaptive rationality raises a host of issues. We address three of these in the context of experimental markets: do consumers differ on the basis of learning; how do these differences, when aggregated, affect market efficiency; and how do consumers learn? Analysis of our experimental data reveals the following. First, multiple segments of consumers exist on the basis of learning. Second, the largest segment consists of subjects who do not learn despite timely feedback and motivation. Third, although some consumers do learn to make optimal choices, the effect of this segment on market efficiency is cancelled by an equal number of subjects who ‘learn' false relations. Finally, although subjects do not learn strict rationality even with experience, they are in the aggregate not so irrational as to allow highly suboptimal brands to survive. Further analysis of how consumers learn, specifically on the cues (signals) and the rules consumers employ in making choices over time leads to the following two conclusions. First, some signals make learning more easy than others: for example, providing market share information improves learning but not as much as providing quality information does. Second, people employ different rules depending upon the type of information they have. For example, consumers making decisions based only on price information are more likely to use a heuristic like ‘buy a medium‐priced product provided it has not failed in the past'. Consumers making decisions based on price and quality information may employ a heuristic such as ‘buy top quality products regardless of price'. We discuss the implications of these findings for theory and practice. Copyright © 2000 John Wiley & Sons, Ltd.     

Instantiated features and the use of "rules."

Classification "rules" in expert and everyday discourse are usually deficient by formal standards, lacking explicit decision procedures and precise terms. The authors argue that a central function of such weak rules is to focus on perceptual learning rather than to provide definitions. In 5 experiments, transfer following learning of family resemblance categories was influenced more by familiar-appearing features than by novel-appearing features equally acceptable under the rule. This occurred both when rules were induced and when rules were given at the beginning of instruction. To model this and other phenomena in categorization, features must be represented on 2 levels: informational and instantiated. These 2 feature levels are crucial to provide broad generalization while reflecting the known peculiarities of a complex world.     

The Neurodynamics of Cognition: A Tutorial on Computational Cognitive Neuroscience

Computational Cognitive Neuroscience (CCN) is a new field that lies at the intersection of computational neuroscience, machine learning, and neural network theory (i.e., connectionism). The ideal CCN model should not make any assumptions that are known to contradict the current neuroscience literature and at the same time provide good accounts of behavior and at least some neuroscience data (e.g., single-neuron activity, fMRI data). Furthermore, once set, the architecture of the CCN network and the models of each individual unit should remain fixed throughout all applications. Because of the greater weight they place on biological accuracy, CCN models differ substantially from traditional neural network models in how each individual unit is modeled, how learning is modeled, and how behavior is generated from the network. A variety of CCN solutions to these three problems are described. A real example of this approach is described, and some advantages and limitations of the CCN approach are discussed.     

Transformation rules in concept learning

Two major classes of models have been proposed to explain concept learning: strength models and distance models (Hayes-Roth & Hayes-Roth, 1977). The present study demonstrates that subjects abstract transformation rules as defined by the Franks and Bransford 11971) distance model. Transformation rules characterize how the patterns of a concept differ from each other. Transformation rules are inconsistent with strength models, which assume that subjects abstract component features and not relational information characterizing the differences among patterns. Whether a strength model or a distance model is more appropriate in other instances of concept learning is probably a function of task demands, stimulus characteristics, and subject characteristics.     

How children perceive fractals: Hierarchical self-similarity and cognitive development

The ability to understand and generate hierarchical structures is a crucial component of human cognition, available in language, music, mathematics and problem solving. Recursion is a particularly useful mechanism for generating complex hierarchies by means of self-embedding rules. In the visual domain, fractals are recursive structures in which simple transformation rules generate hierarchies of infinite depth. Research on how children acquire these rules can provide valuable insight into the cognitive requirements and learning constraints of recursion.     

Implicit and Explicit Learning in Aged and Young Adults

There is a wealth of evidence that learning ability declines with age. In almost all of the studies however, the performance measures employed are explicit, even though research has consistently indicated that aged adults have well preserved implicit learning ability. This suggests that under certain circumstances aged adults should be at no great learning disadvantage in comparison to young adults. This experiment used the artificial grammar-learning paradigm, developed by Reber, in a 2 X 2 factorial design that involved systematic manipulation of grammatical complexity and rule provision. The study explored how each combination of conditions influenced explicit or implicit learning and the relationship between learning style and performance by aged and young adults. Learning was assessed primarily by recognition accuracy, involving correct and incorrect grammar exemplars. However, reaction time, error pattern, reliability, and verbal report data was also collected as a way of confirming and providing added detail on learning performance patterns. Aged adults, irrespective of experimental learning conditions, evidenced a remarkably consistent reaction time deficit. In contrast, the accuracy differential between age groups varied markedly across the four treatments. The most salient contrast occurred between complex grammar, without rules, where there was no difference in accuracy between the two age groups and simple grammar, with rules, where the difference was greatest. This change in learning performance between these two conditions indicates that aged adults will learn as well as young adults in situations where the knowledge domain is conducive to implicit learning.     

Supervised Machine Learning: A Brief Primer

Machine learning is increasingly used in mental health research and has the potential to advance our understanding of how to characterize, predict, and treat mental disorders and associated adverse health outcomes (e.g., suicidal behavior). Machine learning offers new tools to overcome challenges for which traditional statistical methods are not well-suited. This paper provides an overview of machine learning with a specific focus on supervised learning (i.e., methods that are designed to predict or classify an outcome of interest). Several common supervised learning methods are described, along with applied examples from the published literature. We also provide an overview of supervised learning model building, validation, and performance evaluation. Finally, challenges in creating robust and generalizable machine learning algorithms are discussed.     

Implicit sequence learning is represented by stimulus—response rules

For nearly two decades, researchers have investigated spatial sequence learning in an attempt to identify what specifically is learned during sequential tasks (e.g., stimulus order, response order, etc.). Despite extensive research, controversy remains concerning the information-processing locus of this learning effect. There are three main theories concerning the nature of spatial sequence learning, corresponding to the perceptual, motor, or response selection (i.e., central mechanisms underlying the association between stimulus and response pairs) processes required for successful task performance. The present data investigate this controversy and support the theory that stimulus—response (S—R) rules are critical for sequence learning. The results from two experiments demonstrate that sequence learning is disrupted only when the S—R rules for the task are altered. When the S—R rules remain constant or involve only a minor transformation, significant sequence learning occurs. These data implicate spatial response selection as a likely mechanism mediating spatial sequential learning.     

Log Data Analysis with ANFIS: A Fuzzy Neural Network Approach

This study explores the use of the Adaptive Neuro-Fuzzy Inference System (ANFIS), a neuro-fuzzy approach, to analyze the log data of technology-based assessments to extract relevant features of student problem-solving processes, and develop and refine a set of fuzzy logic rules that could be used to interpret student performance. The log data that record student response processes while solving a science simulation task were analyzed with ANFIS. Results indicate the ANFIS analysis could generate and refine a set of fuzzy rules that shed lights on the process of how students solve the simulation task. We conclude the article by discussing the advantages of combining human judgments with the learning capacity of ANFIS for log data analysis and outlining the limitations of the current study and areas of future research.     

Human vs. supervised machine learning: Who learns patterns faster?

The capabilities of supervised machine learning (SML), especially compared to human abilities, are being discussed in scientific research and in the usage of SML. This study provides an answer to how learning performance differs between humans and machines when there is limited training data. We have designed an experiment in which 44 humans and three different machine learning algorithms identify patterns in labeled training data and have to label instances according to the patterns they find. The results show a high dependency between performance and the underlying patterns of the task. Whereas humans perform relatively similarly across all patterns, machines show large performance differences for the various patterns in our experiment. After seeing 20 instances in the experiment, human performance does not improve anymore, which we relate to theories of cognitive overload. Machines learn slower but can reach the same level or may even outperform humans in 2 of the 4 of used patterns. However, machines need more instances compared to humans for the same results. The performance of machines is comparably lower for the other 2 patterns due to the difficulty of combining input features.