On the deleterious effects of a priori objectives on evolution and representation

By Brian G. Woolley, Kenneth O. Stanley. 2011

75 citations recorded by the source corpus.

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

Citations in the corpus

Diversity is All You Need: Learning Skills without a Reward Function (2018, background): “Prior work in neuroevolution and evolutionary algorithms has studied how complex behaviors can be learned by directly maximizing diversity (Lehman & Stanley, 2011a;b; Woolley & Stanley, 2011; Stanley & Miikkulainen, 2002; Such et al., 2017; Pugh et al., 2016; Mouret & Doncieux, 2009).”
Unshackling evolution: evolving soft robots with multiple materials and a powerful generative encoding (2013, methodology): “While our research produced an impressive array of diverse forms, it did use a target-based fitness objective, which can hinder search [38].”
POET: open-ended coevolution of environments and their optimized solutions (2019, background): “Beginning with the introduction of novelty search [38], we have seen in the field a steady loss of confidence in the idea that we can reliably set objectives, or even know what the right objectives are, and pursue them to ambitious ends [65].”
Deep Curiosity Search: Intra-Life Exploration Improves Performance on Challenging Deep Reinforcement Learning Problems (2018, background): “Designing reward functions or state representations to avoid such pitfalls can be difficult [5, 23], especially on any complex (often deceptive) task [24, 25].”
Quality Diversity: A New Frontier for Evolutionary Computation (2016, background): “…Soltoggio and Jones, 2009; Doucette and Heywood, 2010; Goldsby and Cheng, 2010; Graening et al., 2010; Krcah, 2010; Kistemaker and Whiteson, 2011; Woolley and Stanley, 2011; Gomes and Christensen, 2013; Gomes et al., 2013; Liapis et al., 2013a,b; Martinez et al., 2013; Morse et al., 2013;…”
Critical factors in the performance of hyperNEAT (2013, background): “A caveat of our results in Section 6 is that, since each target weights task requires a specific phenotype to solve, they consider only tight fitness functions like those in [28].”
AI-GAs: AI-generating algorithms, an alternate paradigm for producing general artificial intelligence (2019, background): “…ambitious problems, even if you have a way to quantiﬁably measure what success looks like, optimizing for that score is unlikely to produce the correct path through the search space to ﬁnd the global optimum, and instead usually results in getting trapped on a local optima [166, 92, 196, 123].”