{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,4,30]],"date-time":"2025-04-30T04:22:17Z","timestamp":1745986937832,"version":"3.40.4"},"reference-count":33,"publisher":"MIT Press","issue":"1","content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2025,3,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Evolutionary Computation (EC) often throws away learned knowledge as it is reset for each new problem addressed. Conversely, humans can learn from small-scale problems, retain this knowledge (plus functionality), and then successfully reuse them in larger-scale and\/or related problems. Linking solutions to problems has been achieved through layered learning, where an experimenter sets a series of simpler related problems to solve a more complex task. Recent works on Learning Classifier Systems (LCSs) has shown that knowledge reuse through the adoption of Code Fragments, GP-like tree-based programs, is plausible. However, random reuse is inefficient. Thus, the research question is how LCS can adopt a layered-learning framework, such that increasingly complex problems can be solved efficiently. An LCS (named XCSCF*) has been developed to include the required base axioms necessary for learning, refined methods for transfer learning and learning recast as a decomposition into a series of subordinate problems. These subordinate problems can be set as a curriculum by a teacher, but this does not mean that an agent can learn from it; especially if it only extracts over-fitted knowledge of each problem rather than the underlying scalable patterns and functions. Results show that from a conventional tabula rasa, with only a vague notion of which subordinate problems might be relevant, XCSCF* captures the general logic behind the tested domains and therefore can solve any n-bit Multiplexer, n-bit Carry-one, n-bit Majority-on, and n-bit Even-parity problems. This work demonstrates a step towards continual learning as learned knowledge is effectively reused in subsequent problems.<\/jats:p>","DOI":"10.1162\/evco_a_00351","type":"journal-article","created":{"date-parts":[[2024,5,7]],"date-time":"2024-05-07T17:32:33Z","timestamp":1715103153000},"page":"115-140","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":0,"title":["A Layered Learning Approach to Scaling in Learning Classifier Systems for Boolean Problems"],"prefix":"10.1162","volume":"33","author":[{"given":"Isidro M.","family":"Alvarez","sequence":"first","affiliation":[{"name":"School of Engineering and Computer Science, Victoria University of Wellington, Kelburn, Wellington 6140, New Zealand yummyhumans@gmail.com"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1990-8647","authenticated-orcid":true,"given":"Trung B.","family":"Nguyen","sequence":"additional","affiliation":[{"name":"School of Engineering and Computer Science, Victoria University of Wellington, Kelburn, Wellington 6140, New Zealand trung.nguyen@auckland.ac.nz"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8979-2224","authenticated-orcid":true,"given":"Will N.","family":"Browne","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Robotics, Queensland University of Technology, Brisbane 4001, Australia will.browne@qut.edu.au"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4463-9538","authenticated-orcid":true,"given":"Mengjie","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Engineering and Computer Science, Victoria University of Wellington, Kelburn, Wellington 6140, New Zealand mengjie.zhang@ecs.vuw.ac.nz"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","published-online":{"date-parts":[[2025,3,15]]},"reference":[{"key":"2025042916160059000_B1","doi-asserted-by":"crossref","first-page":"969","DOI":"10.1145\/2598394.2611383","article-title":"Reusing learned functionality in XCS: Code fragments with constructed functionality and constructed features","author":"Alvarez","year":"2014","journal-title":"Proceedings of the Companion Publication of the 2014 Annual Conference on Genetic and Evolutionary Computation (GECCO)"},{"key":"2025042916160059000_B2","doi-asserted-by":"crossref","first-page":"383","DOI":"10.1007\/978-3-319-13563-2_33","article-title":"Reusing learned functionality to address complex Boolean functions","author":"Alvarez","year":"2014","journal-title":"Simulated Evolution and Learning"},{"key":"2025042916160059000_B3","first-page":"429","article-title":"Human-inspired scaling in learning classifier systems: Case study on the n-bit multiplexer problem set","author":"Alvarez","year":"2016","journal-title":"Proceedings of the Genetic and Evolutionary Computation Conference (GECCO)"},{"key":"2025042916160059000_B4","first-page":"759","article-title":"A novel multi-task genetic programming approach to uncertain capacitated arc routing problem","author":"Ardeh","year":"2021","journal-title":"Association for Computing Machinery"},{"key":"2025042916160059000_B5","doi-asserted-by":"crossref","first-page":"1630","DOI":"10.1145\/3449726.3463198","article-title":"A genetic fuzzy system for interpretable and parsimonious reinforcement learning policies","author":"Bishop","year":"2021","journal-title":"Proceedings of the Genetic and Evolutionary Computation Conference Companion (GECCO)"},{"issue":"2\u20133","key":"2025042916160059000_B6","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1007\/s12065-015-0125-y","article-title":"A brief history of learning classifier systems: From CS-1 to XCS and its variants","volume":"8","author":"Bull","year":"2015","journal-title":"Evolutionary Intelligence"},{"issue":"3","key":"2025042916160059000_B7","doi-asserted-by":"publisher","first-page":"144","DOI":"10.1007\/s005000100111","article-title":"An algorithmic description of XCS","volume":"6","author":"Butz","year":"2002","journal-title":"Soft Computing"},{"issue":"1","key":"2025042916160059000_B8","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1023\/A:1007379606734","article-title":"Multitask learning","volume":"28","author":"Caruana","year":"1997","journal-title":"Machine Learning"},{"key":"2025042916160059000_B9","first-page":"39","article-title":"Computer science education: The first threshold concept","author":"Falkner","year":"2013","journal-title":"IEEE Learning and Teaching in Computing and Engineering"},{"volume-title":"Adaptation in natural and artificial systems: An introductory analysis with applications to biology, control, and artificial intelligence","year":"1975","author":"Holland","key":"2025042916160059000_B10"},{"key":"2025042916160059000_B11","doi-asserted-by":"crossref","first-page":"263","DOI":"10.1016\/B978-0-12-543104-0.50012-3","article-title":"Adaptation","author":"Holland","year":"1976","journal-title":"Progress in Theoretical Biology"},{"key":"2025042916160059000_B12","first-page":"158","article-title":"Finding general solutions to the parity problem by evolving machine-language representations","author":"Huelsbergen","year":"1998","journal-title":"Genetic Programming"},{"key":"2025042916160059000_B13","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1007\/978-3-540-88138-4_3","article-title":"Investigating scaling of an abstracted LCS utilising ternary and S-expression alphabets","volume-title":"Learning Classifer Systems","author":"Ioannides","year":"2008"},{"key":"2025042916160059000_B14","doi-asserted-by":"crossref","first-page":"350","DOI":"10.1007\/978-3-642-35101-3_30","article-title":"XCSR with computed continuous action","volume-title":"AI 2012: Advances in artificial intelligence","author":"Iqbal","year":"2012"},{"issue":"3","key":"2025042916160059000_B15","doi-asserted-by":"publisher","first-page":"503","DOI":"10.1007\/s00500-012-0922-5","article-title":"Evolving optimum populations with XCS classifier systems","volume":"17","author":"Iqbal","year":"2013","journal-title":"Soft Computing"},{"key":"2025042916160059000_B16","doi-asserted-by":"crossref","first-page":"1045","DOI":"10.1145\/2463372.2463500","article-title":"Extending learning classifier system with cyclic graphs for scalability on complex, large-scale Boolean problems","author":"Iqbal","year":"2013","journal-title":"Proceedings of the 15th Annual Conference on Genetic and Evolutionary Computation (GECCO)"},{"key":"2025042916160059000_B17","doi-asserted-by":"crossref","first-page":"1818","DOI":"10.1109\/CEC.2013.6557781","article-title":"Learning overlapping natured and niche imbalance Boolean problems using XCS classifier systems","author":"Iqbal","year":"2013","journal-title":"2013 IEEE Congress on Evolutionary Computation"},{"issue":"4","key":"2025042916160059000_B18","doi-asserted-by":"publisher","first-page":"465","DOI":"10.1109\/TEVC.2013.2281537","article-title":"Reusing building blocks of extracted knowledge to solve complex, large-scale Boolean problems","volume":"18","author":"Iqbal","year":"2014","journal-title":"IEEE Transactions on Evolutionary Computation"},{"key":"2025042916160059000_B19","first-page":"171","article-title":"A hierarchical approach to learning the Boolean multiplexer function","volume-title":"Foundations of genetic algorithms","author":"Koza","year":"1991"},{"key":"2025042916160059000_B20","doi-asserted-by":"crossref","DOI":"10.4324\/9780203966273","volume-title":"Overcoming barriers to student understanding: Threshold concepts and troublesome knowledge","author":"Meyer","year":"2006"},{"key":"2025042916160059000_B21","doi-asserted-by":"publisher","first-page":"199","DOI":"10.1162\/evco.1995.3.2.199","article-title":"Strongly typed genetic programming","volume":"3","author":"Montana","year":"1995","journal-title":"Evolutionary Computation"},{"key":"2025042916160059000_B22","doi-asserted-by":"crossref","first-page":"428","DOI":"10.1145\/3321707.3321751","article-title":"Improvement of code fragment fitness to guide feature construction in XCS","author":"Nguyen","year":"2019","journal-title":"Proceedings of the Genetic and Evolutionary Computation Conference (GECCO)"},{"key":"2025042916160059000_B23","doi-asserted-by":"crossref","first-page":"3308","DOI":"10.1109\/CEC.2019.8789950","article-title":"Online feature-generation of code fragments for XCS to guide feature construction","author":"Nguyen","year":"2019","journal-title":"2019 IEEE Congress on Evolutionary Computation"},{"issue":"10","key":"2025042916160059000_B24","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"2025042916160059000_B25","article-title":"Learning agile robotic locomotion skills by imitating animals","author":"Peng","year":"2020","journal-title":"Robotics: Science and Systems"},{"key":"2025042916160059000_B26","first-page":"249","article-title":"A cooperative coevolutionary approach to function optimization","author":"Potter","year":"1994","journal-title":"Parallel Problem Solving from Nature"},{"issue":"3-4","key":"2025042916160059000_B27","doi-asserted-by":"publisher","first-page":"262","DOI":"10.1080\/02643290442000095","article-title":"Functional ontologies for cognition: The systematic definition of structure and function","volume":"22","author":"Price","year":"2005","journal-title":"Cognitive Neuropsychology"},{"key":"2025042916160059000_B28","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1007\/3-540-45164-1_38","article-title":"Layered learning","volume-title":"Machine learning: ECML 2000","author":"Stone","year":"2000"},{"key":"2025042916160059000_B29","article-title":"Guided curriculum learning for walking over complex terrain","author":"Tidd","year":"2020","journal-title":"Proceedings of the Australasian Conference on Robotics and Automation"},{"key":"2025042916160059000_B30","first-page":"927","article-title":"Instance-linked attribute tracking and feedback for Michigan-style supervised learning classifier systems","author":"Urbanowicz","year":"2012","journal-title":"Proceedings of the 14th Annual Conference on Genetic and Evolutionary Computation (GECCO)"},{"key":"2025042916160059000_B31","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-662-55007-6","volume-title":"Introduction to learning classifier systems","author":"Urbanowicz","year":"2017"},{"issue":"1","key":"2025042916160059000_B32","doi-asserted-by":"publisher","DOI":"10.1155\/2009\/736398","article-title":"Learning classifier systems: A complete introduction, review, and roadmap","volume":"2009","author":"Urbanowicz","year":"2009","journal-title":"Journal of Artificial Evolution and Applications"},{"issue":"2","key":"2025042916160059000_B33","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1162\/evco.1995.3.2.149","article-title":"Classifier fitness based on accuracy","volume":"3","author":"Wilson","year":"1995","journal-title":"Evolutionary Computation"}],"container-title":["Evolutionary Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/evco\/article-pdf\/33\/1\/115\/2464263\/evco_a_00351.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/evco\/article-pdf\/33\/1\/115\/2464263\/evco_a_00351.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,29]],"date-time":"2025-04-29T20:16:18Z","timestamp":1745957778000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/evco\/article\/33\/1\/115\/120932\/A-Layered-Learning-Approach-to-Scaling-in-Learning"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":33,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,3,15]]},"published-print":{"date-parts":[[2025,3,15]]}},"URL":"https:\/\/doi.org\/10.1162\/evco_a_00351","relation":{},"ISSN":["1530-9304"],"issn-type":[{"type":"electronic","value":"1530-9304"}],"subject":[],"published-other":{"date-parts":[[2025]]},"published":{"date-parts":[[2025]]}}}