Revising the evolutionary computation abstraction: minimal criteria novelty search

By Joel Lehman, Kenneth O. Stanley. 2010

176 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Genetics, Sociology and Political Science

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Genetics

In Sociology and Political Science

Citations in the corpus

Learning quadrupedal locomotion over challenging terrain (2020, methodology): “The method uses minimal criteria (24, 40) and aims to choose environmental parameters that are neither too challenging nor trivial for the agents: This is realized by selecting task parameters that yield midrange rewards.”
Minimal criterion coevolution: a new approach to open-ended search (2017, background): “The main contribution of this paper is to show that coevolution may help to address this problem by forcing the MCs of two interlocking populations to satisfy minimal criteria with respect to each other, even as both populations are gradually shifting simultaneously. idea, called minimal criterion c”
Abandoning Objectives: Evolution Through the Search for Novelty Alone (2011, methodology): “Recently, Lehman and Stanley (2010b) have begun investigating this research direction.”
Illuminating search spaces by mapping elites (2015, background): “Because Nov-elty Search does not work well with very large feature/behavioral spaces 27,28 , there have been many proposals for combining selection for novelty and performance 28–31 .”
Identifying Necessary Conditions for Open-Ended Evolution through the Artificial Life World of Chromaria (2014, background): “Inter-estingly, in evolutionary computation, with a few exceptions (Lehman and Stanley, 2010), most EAs implement no such MC: usually all individuals have at least some small probability of reproduction (De Jong, 2006).”
Quality Diversity: A New Frontier for Evolutionary Computation (2016, methodology): “…sparked considerable research interest in applying NS and methods like it to solving problems that were previously considered to be too difficult (Lehman and Stanley, 2008, 2010, 2011a,b; Kistemaker and Whiteson, 2011; Mouret, 2011; Risi et al., 2011; Mouret and Doncieux, 2012; Cully and Mouret,…”
Evolving plastic neural networks with novelty search (2010, methodology): “Fitness-based NEAT on the other hand performs increasingly poorly as domain deception increases, adding to the growing body of evidence (Lehman & Stanley, 2008, 2010b, 2010c; Mouret, 2009) that novelty search can overcome the deception inherent in a diversity of tasks.”