Pub Date : 2023-10-31DOI: 10.1080/0952813x.2023.2270995
Philipp Seitz, Jan Schmitt
In nonlinear regression with machine learning methods, neural networks (NNs) are ideally suited due to their universal approximation property, which states that arbitrary nonlinear functions can thereby be approximated arbitrarily well. Unfortunately, this property also poses the problem that data points with measurement errors can be approximated too well and unknown parameter subspaces in the estimation can deviate far from the actual value (so-called overfitting). Various developed methods aim to reduce overfitting through modifications in several areas of the training. In this work, we pursue the question of how an NN behaves in training with respect to overfitting when linear and nonlinear transfer functions (TF) are alternated in different hidden layers (HL). The presented approach is applied to a generated dataset and contrasted to established methods from the literature, both individually and in combination. Comparable results are obtained, whereby the common use of purely nonlinear transfer functions proves to be not recommended generally.
{"title":"Alternating Transfer Functions to Prevent Overfitting in Non-Linear Regression with Neural Networks","authors":"Philipp Seitz, Jan Schmitt","doi":"10.1080/0952813x.2023.2270995","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2270995","url":null,"abstract":"In nonlinear regression with machine learning methods, neural networks (NNs) are ideally suited due to their universal approximation property, which states that arbitrary nonlinear functions can thereby be approximated arbitrarily well. Unfortunately, this property also poses the problem that data points with measurement errors can be approximated too well and unknown parameter subspaces in the estimation can deviate far from the actual value (so-called overfitting). Various developed methods aim to reduce overfitting through modifications in several areas of the training. In this work, we pursue the question of how an NN behaves in training with respect to overfitting when linear and nonlinear transfer functions (TF) are alternated in different hidden layers (HL). The presented approach is applied to a generated dataset and contrasted to established methods from the literature, both individually and in combination. Comparable results are obtained, whereby the common use of purely nonlinear transfer functions proves to be not recommended generally.","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"885 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135814206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ABSTRACTThis paper aims to study people’s trust in the capabilities of social robots in the context of digital transformation. Firstly, the current application status of social robots is studied. Then, the capability trust problem of social robots is studied according to the existing problems and phenomena. Students from a municipal experimental middle school are selected for a questionnaire survey, and the anthropomorphism of social robots is taken as the independent variable. The role of social robots with different anthropomorphic degrees on students’ initial capability trust and the mediating role of attraction perception are studied. A research model is established, and SPSS 26.0 is used to further analyse the data. The results show that among the students with a low degree of an anthropomorphic social robot, the average score of anthropomorphism is 2.52, the average score of attraction perception is 3.29, and the score of capability trust is 3.64, which is the upper-middle level. There are significant differences in the initial capability trust evaluation of social robots among students of different ages (F = 38.13, P = 0.00). When the degree of anthropomorphism of social robots is at different levels, there are significant differences in students’ initial capability trust evaluation (F = 34.25, P = 0.00). It can be seen that the degree of anthropomorphism of social robots has an impact on students’ initial capability trust.KEYWORDS: Digital transformationsocial robotsdegree of anthropomorphismcapability trustattraction perception Disclosure statementNo potential conflict of interest was reported by the author(s).Data sharing agreementThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.Additional informationFundingThe author(s) received no financial support for the research, authorship, and/or publication of this article.
摘要本文旨在研究数字化转型背景下人们对社交机器人能力的信任。首先，研究了社交机器人的应用现状。然后，根据存在的问题和现象，对社交机器人的能力信任问题进行了研究。选取某市实验中学的学生进行问卷调查，以社交机器人的拟人化为自变量。研究了不同拟人化程度的社交机器人对学生初始能力信任的影响以及吸引感知的中介作用。建立研究模型，运用SPSS 26.0软件对数据进行进一步分析。结果表明:拟人化程度较低的学生，拟人化平均得分为2.52分，吸引力感知平均得分为3.29分，能力信任得分为3.64分，属于中上水平;不同年龄学生对社交机器人的初始能力信任评价存在显著差异(F = 38.13, P = 0.00)。当社交机器人拟人化程度处于不同水平时，学生的初始能力信任评价存在显著差异(F = 34.25, P = 0.00)。由此可见，社交机器人的拟人化程度对学生的初始能力信任有影响。关键词:数字化转型;社会机器人;拟人化程度;能力;信任;数据共享协议当前研究中使用和/或分析的数据集可根据通讯作者的合理要求提供。作者在研究、撰写和/或发表这篇文章时没有得到任何经济支持。
{"title":"The future of artificial intelligence and digital development: a study of trust in social robot capabilities","authors":"Chuntao Jiang, Xin Guan, Junfan Zhu, Zeyu Wang, Fanbao Xie, Weijia Wang","doi":"10.1080/0952813x.2023.2263456","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2263456","url":null,"abstract":"ABSTRACTThis paper aims to study people’s trust in the capabilities of social robots in the context of digital transformation. Firstly, the current application status of social robots is studied. Then, the capability trust problem of social robots is studied according to the existing problems and phenomena. Students from a municipal experimental middle school are selected for a questionnaire survey, and the anthropomorphism of social robots is taken as the independent variable. The role of social robots with different anthropomorphic degrees on students’ initial capability trust and the mediating role of attraction perception are studied. A research model is established, and SPSS 26.0 is used to further analyse the data. The results show that among the students with a low degree of an anthropomorphic social robot, the average score of anthropomorphism is 2.52, the average score of attraction perception is 3.29, and the score of capability trust is 3.64, which is the upper-middle level. There are significant differences in the initial capability trust evaluation of social robots among students of different ages (F = 38.13, P = 0.00). When the degree of anthropomorphism of social robots is at different levels, there are significant differences in students’ initial capability trust evaluation (F = 34.25, P = 0.00). It can be seen that the degree of anthropomorphism of social robots has an impact on students’ initial capability trust.KEYWORDS: Digital transformationsocial robotsdegree of anthropomorphismcapability trustattraction perception Disclosure statementNo potential conflict of interest was reported by the author(s).Data sharing agreementThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.Additional informationFundingThe author(s) received no financial support for the research, authorship, and/or publication of this article.","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135386759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-21DOI: 10.1080/0952813x.2023.2259913
O. Dalkılıç, N. Demirtaş
ABSTRACTIn this paper, the parameterisation tool of soft set theory is focused and factor sets are defined for all factors that can affect each parameter. Thus, more ideal results are aimed by determining the membership values of the parameters in uncertain environments. In addition, some new hybrid types of soft sets have been defined. The most important advantage of these new hybrid mathematical tools is that they can reduce the possible error margin of decision-makers. Moreover, a decision-making algorithm has been proposed for the set type that can bring us to the most comprehensive data on uncertainty. Finally, the solution to an uncertainty problem is obtained by using the given algorithm.KEYWORDS: Soft setfuzzy soft setuncertainty problemsalgorithmdecision making AcknowledgementsThe authors would like to thank to Mersin University-BAP.Disclosure statementThis article does not contain any studies with human participants or animals performed by any of the authors. Informed consent was obtained from all individual participants included in the paper, including their legal guardians.Additional informationFundingThis study is supported by Mersin University as scientific research project (BAP) with the project code 2022-2-TP3-4769.
{"title":"Novel hybrid soft set theories focusing on decision-makers by considering the factors affecting the parameters","authors":"O. Dalkılıç, N. Demirtaş","doi":"10.1080/0952813x.2023.2259913","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2259913","url":null,"abstract":"ABSTRACTIn this paper, the parameterisation tool of soft set theory is focused and factor sets are defined for all factors that can affect each parameter. Thus, more ideal results are aimed by determining the membership values of the parameters in uncertain environments. In addition, some new hybrid types of soft sets have been defined. The most important advantage of these new hybrid mathematical tools is that they can reduce the possible error margin of decision-makers. Moreover, a decision-making algorithm has been proposed for the set type that can bring us to the most comprehensive data on uncertainty. Finally, the solution to an uncertainty problem is obtained by using the given algorithm.KEYWORDS: Soft setfuzzy soft setuncertainty problemsalgorithmdecision making AcknowledgementsThe authors would like to thank to Mersin University-BAP.Disclosure statementThis article does not contain any studies with human participants or animals performed by any of the authors. Informed consent was obtained from all individual participants included in the paper, including their legal guardians.Additional informationFundingThis study is supported by Mersin University as scientific research project (BAP) with the project code 2022-2-TP3-4769.","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136235454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-20DOI: 10.1080/0952813x.2023.2243277
T. Senthil Prakash, Annalakshmi M, Siva Prasad Patnayakuni, S. Shibu
ABSTRACTEfficacious forecasting of a solid waste supervision system depends on the prediction accuracy of solid waste generation. Several existing methods on municipal solid waste prediction were suggested previously, but those methods do not accurately predict the solid waste, and also it takes high computation time. To overwhelm these issues, Municipal Solid Waste Prediction using Tree Hierarchical Deep Convolutional Neural Network optimised with Balancing Composite Motion Optimization algorithm (MSWP-THDCNN-BCMOA) is proposed for municipal solid waste prediction. Initially, real-time solid waste prediction data is taken from Quantity of MCC, Landfill, Gardan Garbage and Coconut Shell Report in Tamil Nadu (Chennai), such as Zone-9 (Nungambakkam), Zone 10 (Kodambakkam) and Zone 13 (Adyar). Then the collected solid waste data are pre-processed using morphological filtering and extended empirical wavelet transformation. Then the pre-processed data are given to THDCNN-BCMOA algorithm, which accurately predicts the solid waste as wet waste, dry waste, horticulture waste, and dumping yard for 2025–2035 years. The proposed MSWP-THDCNN-BCMOA method is implemented in Python. Then the proposed MSWP-THDCNN-BCMOA method attains 17.91%, 28.30%, 5.63% and 13.54% higher accuracy, 98.66%, 99.13%, 96.43% and 98.31% lower error rate, 53.003%, 48.44%, 25.69% and 42.42% lower computation time compared with existing methods.KEYWORDS: Morphological filtering and extended empirical wavelet transformationtree hierarchical deep convolutional neural networkbalancing composite motion optimizationmunicipal solid waste prediction Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThe author(s) reported there is no funding associated with the work featured in this article.
{"title":"Municipal Solid Waste Prediction using Tree Hierarchical Deep Convolutional Neural Network Optimized with Balancing Composite Motion Optimization Algorithm","authors":"T. Senthil Prakash, Annalakshmi M, Siva Prasad Patnayakuni, S. Shibu","doi":"10.1080/0952813x.2023.2243277","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2243277","url":null,"abstract":"ABSTRACTEfficacious forecasting of a solid waste supervision system depends on the prediction accuracy of solid waste generation. Several existing methods on municipal solid waste prediction were suggested previously, but those methods do not accurately predict the solid waste, and also it takes high computation time. To overwhelm these issues, Municipal Solid Waste Prediction using Tree Hierarchical Deep Convolutional Neural Network optimised with Balancing Composite Motion Optimization algorithm (MSWP-THDCNN-BCMOA) is proposed for municipal solid waste prediction. Initially, real-time solid waste prediction data is taken from Quantity of MCC, Landfill, Gardan Garbage and Coconut Shell Report in Tamil Nadu (Chennai), such as Zone-9 (Nungambakkam), Zone 10 (Kodambakkam) and Zone 13 (Adyar). Then the collected solid waste data are pre-processed using morphological filtering and extended empirical wavelet transformation. Then the pre-processed data are given to THDCNN-BCMOA algorithm, which accurately predicts the solid waste as wet waste, dry waste, horticulture waste, and dumping yard for 2025–2035 years. The proposed MSWP-THDCNN-BCMOA method is implemented in Python. Then the proposed MSWP-THDCNN-BCMOA method attains 17.91%, 28.30%, 5.63% and 13.54% higher accuracy, 98.66%, 99.13%, 96.43% and 98.31% lower error rate, 53.003%, 48.44%, 25.69% and 42.42% lower computation time compared with existing methods.KEYWORDS: Morphological filtering and extended empirical wavelet transformationtree hierarchical deep convolutional neural networkbalancing composite motion optimizationmunicipal solid waste prediction Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThe author(s) reported there is no funding associated with the work featured in this article.","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136308683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-16DOI: 10.1080/0952813x.2023.2256733
V. Kakulapati, A. Jayanthiladevi
ABSTRACTThe corona virus pandemic has affected millions of people’s work and communication. Millions face a health crisis from SARS-CoV-2, the virus that causes most COVID-19 symptoms. The aim of the proposed research is to contribute towards AI (Artificial Intelligence) by developing a mathematical model for SEIR and SIR through CNN on images of affected people and to analyse the dataset of medical images and healthcare outbreaks from 2019 to 2022 to provide an efficient COVID-19 diagnosis tool. The proposed research uses AI and mathematical modelling to develop a learning platform that analyzes images of affected people using CNN to diagnose COVID-19. The dataset used in this research includes medical images and healthcare outbreaks from 2019 to 2022, which are analysed through the SEIR and SIR mathematical models to provide an efficient and accurate COVID-19 diagnosis tool. The results of this research show that the proposed AI learning method is effective in diagnosing COVID-19 using images of affected individuals. The mathematical model for SEIR and SIR, analysed through CNN, provides accurate and efficient diagnosis of COVID-19. The dataset used in this research also provides valuable insights into the outbreak of COVID-19 and its impact on healthcare systems.KEYWORDS: AIchest x-rayCNNCT scanSARS-CoV2 Disclosure statementNo potential conflict of interest was reported by the author(s).
{"title":"Optimal control strategy for COVID-19 developed using an AI-based learning method","authors":"V. Kakulapati, A. Jayanthiladevi","doi":"10.1080/0952813x.2023.2256733","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2256733","url":null,"abstract":"ABSTRACTThe corona virus pandemic has affected millions of people’s work and communication. Millions face a health crisis from SARS-CoV-2, the virus that causes most COVID-19 symptoms. The aim of the proposed research is to contribute towards AI (Artificial Intelligence) by developing a mathematical model for SEIR and SIR through CNN on images of affected people and to analyse the dataset of medical images and healthcare outbreaks from 2019 to 2022 to provide an efficient COVID-19 diagnosis tool. The proposed research uses AI and mathematical modelling to develop a learning platform that analyzes images of affected people using CNN to diagnose COVID-19. The dataset used in this research includes medical images and healthcare outbreaks from 2019 to 2022, which are analysed through the SEIR and SIR mathematical models to provide an efficient and accurate COVID-19 diagnosis tool. The results of this research show that the proposed AI learning method is effective in diagnosing COVID-19 using images of affected individuals. The mathematical model for SEIR and SIR, analysed through CNN, provides accurate and efficient diagnosis of COVID-19. The dataset used in this research also provides valuable insights into the outbreak of COVID-19 and its impact on healthcare systems.KEYWORDS: AIchest x-rayCNNCT scanSARS-CoV2 Disclosure statementNo potential conflict of interest was reported by the author(s).","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"198 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135306532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-16DOI: 10.1080/0952813x.2023.2256739
Neeraj Nagwanshi, Anjali Potnis
ABSTRACTA familiar nervous system disorder characterised by seizures is called as Epilepsy. It is indeed hard to control the suitable type as an outcome of insufficient EEG information. In order to overcome these issues, a Multilayer Neural Network (MLNN)-based classifier is proposed to recognise if the patients are affected by epileptic disease or not. EEG signal is a contribution, and the input signal is preprocessed using antialiasing filter, finite impulse response, and band pass filter to eradicate unwanted noise present in the signal. After preprocessing, the features extracting process is done, and four extraction techniques are proposed in order to calculate the feature coefficient. The feature extraction outcome is fed into the MLNN classifier to predict the disease. MLNN performs with Coot-Optimization to reduce error and increase prediction accuracy. The future ideal applied in Matlab-software carried out numerous act metrics, and these parameters attained better performance such as accuracy of 96.5%, error of 0.03, precision of 98%, specificity is 97%, sensitivity is 95%, and so on. This displays the effectiveness of the future ideal than existing approaches such as ANN, SVM, KNN and NB. Based on this proposed classification, the epileptic disease prediction can be improved on this technique and can provide a living standard for patients.KEYWORDS: Epilepsy diseaseeeg signalmulti-layer neural networkantialiasing filterfinite impulse response Author contributionsThe corresponding author claims the major contribution of the paper including formulation, analysis and editing. The co-author provide guidance to verify the analysis result and manuscript editing.Compliance with ethical standardsThis article is a completely original work of its authors; it has not been published before and will not be sent to other publications until the journal’s editorial board decides not to accept it for publication.Disclosure statementNo potential conflict of interest was reported by the author(s).FundingThe authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
{"title":"Detection of Epilepsy patients using coot optimization based feed forward multilayer neural network","authors":"Neeraj Nagwanshi, Anjali Potnis","doi":"10.1080/0952813x.2023.2256739","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2256739","url":null,"abstract":"ABSTRACTA familiar nervous system disorder characterised by seizures is called as Epilepsy. It is indeed hard to control the suitable type as an outcome of insufficient EEG information. In order to overcome these issues, a Multilayer Neural Network (MLNN)-based classifier is proposed to recognise if the patients are affected by epileptic disease or not. EEG signal is a contribution, and the input signal is preprocessed using antialiasing filter, finite impulse response, and band pass filter to eradicate unwanted noise present in the signal. After preprocessing, the features extracting process is done, and four extraction techniques are proposed in order to calculate the feature coefficient. The feature extraction outcome is fed into the MLNN classifier to predict the disease. MLNN performs with Coot-Optimization to reduce error and increase prediction accuracy. The future ideal applied in Matlab-software carried out numerous act metrics, and these parameters attained better performance such as accuracy of 96.5%, error of 0.03, precision of 98%, specificity is 97%, sensitivity is 95%, and so on. This displays the effectiveness of the future ideal than existing approaches such as ANN, SVM, KNN and NB. Based on this proposed classification, the epileptic disease prediction can be improved on this technique and can provide a living standard for patients.KEYWORDS: Epilepsy diseaseeeg signalmulti-layer neural networkantialiasing filterfinite impulse response Author contributionsThe corresponding author claims the major contribution of the paper including formulation, analysis and editing. The co-author provide guidance to verify the analysis result and manuscript editing.Compliance with ethical standardsThis article is a completely original work of its authors; it has not been published before and will not be sent to other publications until the journal’s editorial board decides not to accept it for publication.Disclosure statementNo potential conflict of interest was reported by the author(s).FundingThe authors declare that no funds, grants, or other support were received during the preparation of this manuscript.","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135306533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-24DOI: 10.1080/0952813x.2023.2222374
Shanna Dobson, Chris Fields
If episodic memory is constructive, experienced time is also a construct. We develop an event-based formalism that replaces the traditional objective, agent-independent notion of time with a constructive, agent-dependent notion of time. We show how to make this agent-dependent time entropic and hence well-defined. We use sheaf-theoretic techniques to render agent-dependent time functorial and to construct episodic memories as sequences of observed and constructed events with well-defined limits that maximise the consistency of categorisations assigned to objects appearing in memories. We then develop a condensed formalism that represents episodic memories as pure constructs from single events. We formulate an empirical hypothesis that human episodic memory implements a particular time-symmetric constructive functor, and discuss possible experimental tests.
{"title":"Constructing condensed memories in functorial time","authors":"Shanna Dobson, Chris Fields","doi":"10.1080/0952813x.2023.2222374","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2222374","url":null,"abstract":"If episodic memory is constructive, experienced time is also a construct. We develop an event-based formalism that replaces the traditional objective, agent-independent notion of time with a constructive, agent-dependent notion of time. We show how to make this agent-dependent time entropic and hence well-defined. We use sheaf-theoretic techniques to render agent-dependent time functorial and to construct episodic memories as sequences of observed and constructed events with well-defined limits that maximise the consistency of categorisations assigned to objects appearing in memories. We then develop a condensed formalism that represents episodic memories as pure constructs from single events. We formulate an empirical hypothesis that human episodic memory implements a particular time-symmetric constructive functor, and discuss possible experimental tests.","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135842243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-16DOI: 10.1080/0952813x.2023.2184990
This article refers to:Retracted Article: Artificial intelligence for the identification of healthy fruits and vegetables using MMDL-ABO
本文指:撤稿文章:利用MMDL-ABO进行健康果蔬识别的人工智能
{"title":"Retraction","authors":"","doi":"10.1080/0952813x.2023.2184990","DOIUrl":"https://doi.org/10.1080/0952813x.2023.2184990","url":null,"abstract":"This article refers to:Retracted Article: Artificial intelligence for the identification of healthy fruits and vegetables using MMDL-ABO","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135488545","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-10-01DOI: 10.1080/09528138908953711
Aaron Slomon
Abstract This paper contrasts the standard (in AI) ‘modular’ theory of the nature of vision with a more general theory of vision as involving multiple functions and multiple relationships with other sub-systems of an intelligent system. The modular theory (e.g. as expounded by Marr) treats vision as entirely, and permanently, concerned with the production of a limited range of descriptions of visible surfaces, for a central database; while the ‘labyrinthine’ design allows any output that a visual system can be trained to associate reliably with features of an optic array and allows forms of learning that set up new communication channels. The labyrinthine theory turns out to have much in common with J.J. Gibson's theory of affordances, while not eschewing information processing as he did. It also seems to fit better than the modular theory with neurophysiological evidence of rich interconnectivity within and between sub-systems in the brain. Some of the trade-offs between different designs are discussed i...
{"title":"On designing a visual system# (towards a Gibsonian computational model of vision)","authors":"Aaron Slomon","doi":"10.1080/09528138908953711","DOIUrl":"https://doi.org/10.1080/09528138908953711","url":null,"abstract":"Abstract This paper contrasts the standard (in AI) ‘modular’ theory of the nature of vision with a more general theory of vision as involving multiple functions and multiple relationships with other sub-systems of an intelligent system. The modular theory (e.g. as expounded by Marr) treats vision as entirely, and permanently, concerned with the production of a limited range of descriptions of visible surfaces, for a central database; while the ‘labyrinthine’ design allows any output that a visual system can be trained to associate reliably with features of an optic array and allows forms of learning that set up new communication channels. The labyrinthine theory turns out to have much in common with J.J. Gibson's theory of affordances, while not eschewing information processing as he did. It also seems to fit better than the modular theory with neurophysiological evidence of rich interconnectivity within and between sub-systems in the brain. Some of the trade-offs between different designs are discussed i...","PeriodicalId":133720,"journal":{"name":"Journal of Experimental and Theoretical Artificial Intelligence","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114519481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：