دورية أكاديمية
A Layered Learning Approach to Scaling in Learning Classifier Systems for Boolean Problems.
العنوان: | A Layered Learning Approach to Scaling in Learning Classifier Systems for Boolean Problems. |
---|---|
المؤلفون: | Alvarez IM; School of Engineering and Computer Science, Victoria University of Wellington, Kelburn,Wellington 6140, New Zealand yummyhumans@gmail.com., Nguyen TB; School of Engineering and Computer Science, Victoria University of Wellington, Kelburn,Wellington 6140, New Zealand trung.nguyen@auckland.ac.nz., Browne WN; School of Electrical Engineering and Robotics, Queensland University of Technology, Brisbane 4001, Australia will.browne@qut.edu.au., Zhang M; School of Engineering and Computer Science, Victoria University of Wellington, Kelburn,Wellington 6140, New Zealand mengjie.zhang@ecs.vuw.ac.nz. |
المصدر: | Evolutionary computation [Evol Comput] 2024 May 06, pp. 1-25. Date of Electronic Publication: 2024 May 06. |
Publication Model: | Ahead of Print |
نوع المنشور: | Journal Article |
اللغة: | English |
بيانات الدورية: | Publisher: MIT Press Country of Publication: United States NLM ID: 9513581 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1530-9304 (Electronic) Linking ISSN: 10636560 NLM ISO Abbreviation: Evol Comput Subsets: MEDLINE |
أسماء مطبوعة: | Original Publication: Cambridge, Mass. : MIT Press, c1993- |
مستخلص: | 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 together 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 what 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. (© 2024 Massachusetts Institute of Technology.) |
فهرسة مساهمة: | Keywords: Code Fragments; Layered Learning; Learning Classifier Systems |
تواريخ الأحداث: | Date Created: 20240507 Latest Revision: 20240507 |
رمز التحديث: | 20240508 |
DOI: | 10.1162/evco_a_00351 |
PMID: | 38713737 |
قاعدة البيانات: | MEDLINE |
تدمد: | 1530-9304 |
---|---|
DOI: | 10.1162/evco_a_00351 |