Mathias K. Nilsen, Tønnes F. Nygaard, Kai Olav Ellefsen
{"title":"Reward tampering and evolutionary computation: a study of concrete AI-safety problems using evolutionary algorithms","authors":"Mathias K. Nilsen, Tønnes F. Nygaard, Kai Olav Ellefsen","doi":"10.1007/s10710-023-09456-0","DOIUrl":"https://doi.org/10.1007/s10710-023-09456-0","url":null,"abstract":"Abstract Reward tampering is a problem that will impact the trustworthiness of the powerful AI systems of the future. Reward Tampering describes the problem where AI agents bypass their intended objective, enabling unintended and potentially harmful behaviours. This paper investigates whether the creative potential of evolutionary algorithms could help ensure trustworthy solutions when facing this problem. The reason why evolutionary algorithms may help combat reward tampering is that they are able to find a diverse collection of different solutions to a problem within a single run, aiding the search for desirable solutions. Four different evolutionary algorithms were deployed in tasks illustrating the problem of reward tampering. The algorithms were designed with varying degrees of human expertise, measuring how human guidance influences the ability to discover trustworthy solutions. The results indicate that the algorithms’ ability to find and preserve trustworthy solutions is very dependent on preserving diversity during the search. Algorithms searching for behavioural diversity showed to be the most effective against reward tampering. Human expertise also showed to improve the certainty and quality of safe solutions, but even with only a minimal degree of human expertise, domain-independent diversity management was found to discover safe solutions.","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135061119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Evolutionary design and analysis of ribozyme-based logic gates","authors":"Nicolas Kamel, N. Kharma, Jonathan Perreault","doi":"10.1007/s10710-023-09459-x","DOIUrl":"https://doi.org/10.1007/s10710-023-09459-x","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2023-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43171199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"GAAMmf: genetic algorithm with aggressive mutation and decreasing feature set for feature selection","authors":"I. Rejer, Krzysztof Lorenz","doi":"10.1007/s10710-023-09458-y","DOIUrl":"https://doi.org/10.1007/s10710-023-09458-y","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"24 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2023-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41491921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Marlen Meza-Sánchez, M. C. R. Liñán, Eddie Clemente, Leonardo Herrera
{"title":"Evolutionary design of swing-up controllers for stabilization task of underactuated inverted pendulums","authors":"Marlen Meza-Sánchez, M. C. R. Liñán, Eddie Clemente, Leonardo Herrera","doi":"10.1007/s10710-023-09457-z","DOIUrl":"https://doi.org/10.1007/s10710-023-09457-z","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2023-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49034878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Fall compensation detection from EEG using neuroevolution and genetic hyperparameter optimisation","authors":"Jordan J. Bird, Ahmad Lotfi","doi":"10.1007/s10710-023-09453-3","DOIUrl":"https://doi.org/10.1007/s10710-023-09453-3","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"24 1","pages":"1-26"},"PeriodicalIF":2.6,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44708692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A comparison of an evolvable hardware controller with an artificial neural network used for evolving the gait of a hexapod robot","authors":"Fraser Borrett, M. Beckerleg","doi":"10.1007/s10710-023-09452-4","DOIUrl":"https://doi.org/10.1007/s10710-023-09452-4","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"24 1","pages":"1-30"},"PeriodicalIF":2.6,"publicationDate":"2023-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45567326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Tasos Asonitis, R. Allmendinger, Matt Benatan, R. Climent
{"title":"SonOpt: understanding the behaviour of bi-objective population-based optimisation algorithms through sound","authors":"Tasos Asonitis, R. Allmendinger, Matt Benatan, R. Climent","doi":"10.1007/s10710-023-09451-5","DOIUrl":"https://doi.org/10.1007/s10710-023-09451-5","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"24 1","pages":"1-41"},"PeriodicalIF":2.6,"publicationDate":"2023-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46696837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Framework for unsupervised incremental evolution of stylized images","authors":"Florian Uhde","doi":"10.1007/s10710-023-09449-z","DOIUrl":"https://doi.org/10.1007/s10710-023-09449-z","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"24 1","pages":"1-21"},"PeriodicalIF":2.6,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42678251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}