On Synergistic Interactions Between Evolution, Development and Layered Learning

Hoang, Tuan-Hao; McKay, Bob; Essam, Daryl; Hoài, Nguyễn Xuân

doi:10.1109/tevc.2011.2150752

Cited by 22 publications

(19 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although DTAG3P outperformed the two single-short learning GP systems, the standard tree-GP system [24] and the TAG3P system [41], it could not achieve 100% success rate for the 8-bit even-parity problem. The reported success rates are 6.67%, 10%, and 86.67% for the standard tree-GP, the TAG3P, and the DTAG3P systems, respectively [33].…”

Section: B Results Comparison With Gp Systemsmentioning

confidence: 98%

“…The third comparison is with the DTAG3P system developed in [33]. Using the DTAG3P system, the 8-bit even-parity problem was experimented in a layered learning fashion using the function set {and, or, not, xor}, the population size max pop = 250, and the number of maximum generations at each problem level max gen = 101.…”

Section: B Results Comparison With Gp Systemsmentioning

confidence: 99%

“…A number of GP researchers have incorporated and investigated layered learning in GP [31]- [33]. CGP and layered learning GP are briefly introduced here as they are closely related to the work presented in this paper.…”

Section: B Evolutionary Computationmentioning

confidence: 99%

“…Hoang et al [33] investigated interactions between evolution, development, and layered learning using tree adjoining grammar guided GP (TAG3P) [41]. The developed system, called DTAG3P, was tested in symbolic regression problems, Boolean even-parity problems, and ORDERTREE problems.…”

Section: B Evolutionary Computationmentioning

confidence: 99%

See 3 more Smart Citations

Reusing Building Blocks of Extracted Knowledge to Solve Complex, Large-Scale Boolean Problems

Iqbal

Browne

Zhang

2014

IEEE Trans. Evol. Computat.

136

147

View full text Add to dashboard Cite

Evolutionary computation techniques have had limited capabilities in solving large-scale problems due to the large search space demanding large memory and much longer training times. In the work presented here, a genetic programming like rich encoding scheme has been constructed to identify building blocks of knowledge in a learning classifier system. The fitter building blocks from the learning system trained against smaller problems have been utilized in a higher complexity problem in the domain to achieve scalable learning. The proposed system has been examined and evaluated on four different Boolean problem domains: 1) multiplexer, 2) majority-on, 3) carry, and 4) evenparity problems. The major contribution of this paper is to successfully extract useful building blocks from smaller problems and reuse them to learn more complex large-scale problems in the domain, e.g., 135-bit multiplexer problem, where the number of possible instances is 2 135 ≈ 4 × 10 40 , is solved by reusing the extracted knowledge from the learned lower level solutions in the domain. Autonomous scaling is, for the first time, shown to be possible in learning classifier systems. It improves effectiveness and reduces the number of training instances required in large problems, but requires more time due to its sequential build-up of knowledge.

show abstract

Section: B Results Comparison With Gp Systemsmentioning

confidence: 98%

Section: B Results Comparison With Gp Systemsmentioning

confidence: 99%

Section: B Evolutionary Computationmentioning

confidence: 99%

Section: B Evolutionary Computationmentioning

confidence: 99%

See 2 more Smart Citations

Reusing Building Blocks of Extracted Knowledge to Solve Complex, Large-Scale Boolean Problems

Iqbal

Browne

Zhang

2014

IEEE Trans. Evol. Computat.

136

147

View full text Add to dashboard Cite

show abstract

“…To be sure, precise predictions about the future of evo-devo are impossible (Hall, 2012), but the possibility, and in consequence the responsibility, is to an extent in our hands to steer the boat toward intellectually rewarding paths. We are already witnessing our discipline expanding its scope by exploring its boundaries toward other disciplines, ecology in particular (eco-evodevo, e.g., Hall, 2003;Gilbert and Epel, 2008; also evo-devo vs. niche construction, e.g., Laland et al, 2008), but also ethology (e.g., Bertossa, 2011) and the sciences of language (e.g., Hoang et al, 2011;Hall, 2013). Future advances in all these directions are welcome, but I expect that much more progress will come from the impact that evo-devo will eventually have on the foundations and the research agendas of its parent disciplines.…”

Section: Questions To Be Addressedmentioning

confidence: 99%