Improving Genetic Programming with Novel Exploration - Exploitation Control

Kelly, Jonathan; Hemberg, Erik; O’Reilly, Una-May

doi:10.1007/978-3-030-16670-0_5

Cited by 27 publications

(16 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A watershed moment was the introduction of a program synthesis benchmark suite of 29 problems, systematically selected from sources teaching introductory computer science programming [8]. Four studies are most relevant to this paper: the original benchmarking done using PUSHGP [8], the most recent PUSHGP results using various mutation operators [7], the most recent grammar guided GP eorts [4], and, recently introduced, grammatical evolution with knobelty selection [9]. Henceforth we refer to these as PushGP BM , PushGP MU , G3P, and GE N OV respectively.…”

Section: Program Synthesis With Gpmentioning

confidence: 99%

“…Another is knobelty selection [9]. It parametrically balances exploration and exploitation by populating a mixed population of ospring.…”

Section: Selection Operators For Program Synthesismentioning

confidence: 99%

“…This sampling introduces heuristic eciency in lieu of exhaustive correctness and dispenses with a complex cache management policy. A known side eect of novelty search is its tendency to cause solutions to bloat, as solutions that are longer appear to be more novel when, functionally, they are not [9]. Additionally, in GP, tree based operators also have a tendency to induce bloat [16].…”

Section: Selection Operators For Program Synthesismentioning

confidence: 99%

“…More recently, program synthesis has supported the investigation of how tunable selection-based novelty can help balance search exploration and exploitation. Simultaneously, narrow program synthesis performance improvement was also demonstrated [9]. As these are some of the most recent contributions to the program synthesis community, for the rest of the paper we refer to them as state of art.…”

mentioning

confidence: 94%

See 3 more Smart Citations

On domain knowledge and novelty to improve program synthesis performance with grammatical evolution

Hemberg

Kelly

O’Reilly

2019

Proceedings of the Genetic and Evolutionary Computation Conference

Self Cite

View full text Add to dashboard Cite

Programmers solve coding problems with the support of both programming and problem specic knowledge. They integrate this domain knowledge to reason by computational abstraction. Correct and readable code arises from sound abstractions and problem solving. We attempt to transfer insights from such human expertise to genetic programming (GP) for solving automatic program synthesis. We draw upon manual and non-GP Articial Intelligence methods to extract knowledge from synthesis problem denitions to guide the construction of the grammar that Grammatical Evolution uses and to supplement its tness function. We examine the impact of using such knowledge on 21 problems from the GP program synthesis benchmark suite. Additionally, we investigate the compounding impact of this knowledge and novelty search. The resulting approaches exhibit improvements in accuracy on a majority of problems in the eld's benchmark suite of program synthesis problems. CCS CONCEPTS • Computing methodologies → Multi-agent systems;

show abstract

Section: Program Synthesis With Gpmentioning

confidence: 99%

“…Another is knobelty selection [9]. It parametrically balances exploration and exploitation by populating a mixed population of ospring.…”

Section: Selection Operators For Program Synthesismentioning

confidence: 99%

Section: Selection Operators For Program Synthesismentioning

confidence: 99%

mentioning

confidence: 94%

See 2 more Smart Citations

On domain knowledge and novelty to improve program synthesis performance with grammatical evolution

Hemberg

Kelly

O’Reilly

2019

Proceedings of the Genetic and Evolutionary Computation Conference

Self Cite

View full text Add to dashboard Cite

show abstract

“…Perhaps the most similar work to that discussed here is another effort to combine novelty search and objective performance, called knobelty [16]. This work takes a different approach, with each parent selection event choosing to either select based on novelty (using tournament selection) or based on performance (using lexicase selection).…”

Section: Combining Novelty and Lexicasementioning

confidence: 99%

Comparing and combining lexicase selection and novelty search

Jundt

Helmuth

2019

Proceedings of the Genetic and Evolutionary Computation Conference

View full text Add to dashboard Cite

Lexicase selection and novelty search, two parent selection methods used in evolutionary computation, emphasize exploring widely in the search space more than traditional methods such as tournament selection. However, lexicase selection is not explicitly driven to select for novelty in the population, and novelty search suffers from lack of direction toward a goal, especially in unconstrained, highly-dimensional spaces. We combine the strengths of lexicase selection and novelty search by creating a novelty score for each test case, and adding those novelty scores to the normal error values used in lexicase selection. We use this new novelty-lexicase selection to solve automatic program synthesis problems, and find it significantly outperforms both novelty search and lexicase selection. Additionally, we find that novelty search has very little success in the problem domain of program synthesis. We explore the effects of each of these methods on population diversity and long-term problem solving performance, and give evidence to support the hypothesis that novelty-lexicase selection resists converging to local optima better than lexicase selection.

show abstract