Progressive Minimal Criteria Novelty Search

Gomes, Jorge; Urbano, Paulo; Christensen, Anders Lyhne

doi:10.1007/978-3-642-34654-5_29

Cited by 19 publications

(30 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the search process , restarting converged evolutionary runs using novelty search , a minimal criteria (for survival and reproduction of controller behaviors) novelty search (Lehman and Stanley, 2010b), a progressive minimal criteria incrementing the requirements for reproduction throughout the evolutionary process (Gomes et al, 2012) and novelty search combined with speciation techniques (Inden et al, 2013), with the result of yielding optimal and near optimal solutions in various tasks including pole-balancing, maze solving, and quadruped gait evolution tasks.…”

Section: Gns Variant: Genotypic Novelty Searchmentioning

confidence: 99%

“…Gomes et al (2012) found that their progressive minimal criteria novelty metric out-performed pure NS in a swarm robotics task. However, it has also been found that an objective-based search can out-perform NS on the deceptive tartarus task as well as pole-balancing and a visual discrimination task (Inden et al, 2013).…”

Section: Gns Variant: Genotypic Novelty Searchmentioning

confidence: 99%

“…In line with previous research on hybrid NS and fitness metrics supporting performance gains in various tasks , including multi-robot ER tasks (Gomes et al, 2012), the ONS variant uses a behavioral diversity metric that linearly combines NS with the objective-based search (fitness functions) native to HyperNEAT. Thus, we use a linear combination of fitness and novelty scores (Gomes et al, 2014), specified in equation (4):…”

Section: Gns Variant: Genotypic Novelty Searchmentioning

confidence: 99%

See 2 more Smart Citations

Evolutionary Policy Transfer and Search Methods for Boosting Behavior Quality: RoboCup Keep-Away Case Study

Nitschke

Didi

2017

Front. Robot. AI

View full text Add to dashboard Cite

This study evaluates various evolutionary search methods to direct neural controller evolution in company with policy (behavior) transfer across increasingly complex collective robotic (RoboCup keep-away) tasks. Robot behaviors are first evolved in a source task and then transferred for further evolution to more complex target tasks. Evolutionary search methods tested include objective-based search (fitness function), behavioral and genotypic diversity maintenance, and hybrids of such diversity maintenance and objective-based search. Evolved behavior quality is evaluated according to effectiveness and efficiency. Effectiveness is the average task performance of transferred and evolved behaviors, where task performance is the average time the ball is controlled by a keeper team. Efficiency is the average number of generations taken for the fittest evolved behaviors to reach a minimum task performance threshold given policy transfer. Results indicate that policy transfer coupled with hybridized evolution (behavioral diversity maintenance and objective-based search) addresses the bootstrapping problem for increasingly complex keep-away tasks. That is, this hybrid method (coupled with policy transfer) evolves behaviors that could not otherwise be evolved. Also, this hybrid evolutionary search was demonstrated as consistently evolving topologically simple neural controllers that elicited high-quality behaviors.

show abstract

Section: Gns Variant: Genotypic Novelty Searchmentioning

confidence: 99%

Section: Gns Variant: Genotypic Novelty Searchmentioning

confidence: 99%

Section: Gns Variant: Genotypic Novelty Searchmentioning

confidence: 99%

See 1 more Smart Citation

Evolutionary Policy Transfer and Search Methods for Boosting Behavior Quality: RoboCup Keep-Away Case Study

Nitschke

Didi

2017

Front. Robot. AI

View full text Add to dashboard Cite

show abstract

“…Other authors have confirmed that novelty search is a promising alternative to objective-based search. For instance, Risi et al showed that novelty search is capable of circumventing the deceptive trap of "learning to learn" when evolving plastic neural networks [156], Krčah successfully applied it to body-brain co-evolution [100], and Gomes et al to swarm robotics [71].…”

Section: Task-agnosticmentioning

confidence: 99%

“…They concluded that this approach can lead to better results than simple novelty search in a deceptive maze navigation task. This idea has been extended by Gomes et al [71], who proposed to make the criteria dependent on the success of the current population. They illustrated their approach with a swarm robotics task [71].…”

Section: Task-agnosticmentioning

confidence: 99%

Beyond black-box optimization: a review of selective pressures for evolutionary robotics

Doncieux

Mouret

2014

Evol. Intel.

View full text Add to dashboard Cite

Evolutionary robotics is often viewed as the application of a family of black-box optimization algorithms -evolutionary algorithms -to the design of robots, or parts of robots. When considering evolutionary robotics as black-box optimization, the selective pressure is mainly driven by a user-defined, black-box fitness function, and a domain-independent selection procedure. However, most evolutionary robotics experiments face similar challenges in similar setups: the selective pressure, and, in particular, the fitness function, is not a pure user-defined black box. The present review shows that, because evolutionary robotics experiments share common features, selective pressures for evolutionary robotics are a subject of research on their own. The literature has been split into two categories: goal refiners, aimed at changing the definition of a good solution, and process helpers, designed to help the search process. Two subcategories are further considered: task-specific approaches, which require knowledge on how to solve the task and taskagnostic ones, which do not need it. Besides highlighting the diversity of the approaches and their respective goals, the present review shows that many task-agnostic process helpers have been proposed during the last years, thus bringing us closer to the goal of a fully automated robot behavior design process.

show abstract