Fully Autonomous Real-Time Autoencoder-Augmented Hebbian Learning through the Collection of Novel Experiences

Bowren, Joshua; Pugh, Justin K.; Stanley, Kenneth O.

doi:10.7551/978-0-262-33936-0-ch064

Cited by 5 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, several approaches have been proposed to combine novelty with a more traditional fitness objective [17,20,37,38] to reorient search towards fitness as it explores the behavior space. These approaches have helped scale novelty search to more complex environments, including an array of control [3,13,37] and content generation [27,29,30] domains.…”

Section: Novelty Searchmentioning

confidence: 99%

Enhanced optimization with composite objectives and novelty pulsation

Shahrzad

Hodjat

Dollé

et al. 2019

Proceedings of the Genetic and Evolutionary Computation Conference Companion

View full text Add to dashboard Cite

An important benefit of multi-objective search is that it maintains a diverse population of candidates, which helps in deceptive problems in particular. Not all diversity is useful, however: candidates that optimize only one objective while ignoring others are rarely helpful. A recent solution is to replace the original objectives by their linear combinations, thus focusing the search on the most useful tradeoffs between objectives. To compensate for the loss of diversity, this transformation is accompanied by a selection mechanism that favors novelty. This paper improves this approach further by introducing novelty pulsation, i.e. a systematic method to alternate between novelty selection and local optimization. In the highly deceptive problem of discovering minimal sorting networks, it finds state-of-the-art solutions significantly faster than before. In fact, our method so far has established a new world record for the 20-line sorting network with 91 comparators. In the real-world problem of stock trading, it discovers solutions that generalize significantly better on unseen data. Composite Novelty Pulsation is therefore a promising approach to solving deceptive real-world problems through multi-objective optimization.

show abstract

Section: Novelty Searchmentioning

confidence: 99%

Enhanced optimization with composite objectives and novelty pulsation

Shahrzad

Hodjat

Dollé

et al. 2019

Proceedings of the Genetic and Evolutionary Computation Conference Companion

View full text Add to dashboard Cite

show abstract

“…Recently, several approaches have been proposed to combine novelty with a more traditional fitness objective (Gomes et al 2015;Gomes, 2009;Mouret, 2011;Mouret and Doncieux, 2012;Pugh et al 2015) to reorient search towards fitness as it explores the behavior space. These approaches have helped scale novelty search to more complex environments, including an array of control (Bowren et al 2016;Cully et al 2015;Mouret and Doncieux, 2012) and content generation (Lehman et al 2016;Lehman and Stanley, 2012; domains.…”

Section: Novelty Searchmentioning

confidence: 99%

Enhanced Optimization with Composite Objectives and Novelty Selection

Shahrzad¹,

Fink²,

Miikkulainen³

2018

The 2018 Conference on Artificial Life

View full text Add to dashboard Cite

show abstract

“…Recently, several approaches have been proposed to combine novelty with a more traditional tness objective [11,12,25,26,31] to reorient search towards tness as it explores the behavior space. ese approaches have helped scale novelty search to more complex environments, including an array of control [3,6,27] and content generation [16,20,21,23,[28][29][30] domains. is paper shows that, aside from focusing search overall, the addition of tness can also be used to focus search on discovering useful stepping stones.…”

Section: Introductionmentioning

confidence: 99%

Discovering evolutionary stepping stones through behavior domination

Meyerson

Miikkulainen

2017

Proceedings of the Genetic and Evolutionary Computation Conference

View full text Add to dashboard Cite

Behavior domination is proposed as a tool for understanding and harnessing the power of evolutionary systems to discover and exploit useful stepping stones. Novelty search has shown promise in overcoming deception by collecting diverse stepping stones, and several algorithms have been proposed that combine novelty with a more traditional tness measure to refocus search and help novelty search scale to more complex domains. However, combinations of novelty and tness do not necessarily preserve the stepping stone discovery that novelty search a ords. In several existing methods, competition between solutions can lead to an unintended loss of diversity. Behavior domination de nes a class of algorithms that avoid this problem, while inheriting theoretical guarantees from multiobjective optimization. Several existing algorithms are shown to be in this class, and a new algorithm is introduced based on fast non-dominated sorting. Experimental results show that this algorithm outperforms existing approaches in domains that contain useful stepping stones, and its advantage is sustained with scale. e conclusion is that behavior domination can help illuminate the complex dynamics of behavior-driven search, and can thus lead to the design of more scalable and robust algorithms.

show abstract

Fully Autonomous Real-Time Autoencoder-Augmented Hebbian Learning through the Collection of Novel Experiences

Cited by 5 publications

References 0 publications

Enhanced optimization with composite objectives and novelty pulsation

Enhanced optimization with composite objectives and novelty pulsation

Enhanced Optimization with Composite Objectives and Novelty Selection

Discovering evolutionary stepping stones through behavior domination

Contact Info

Product

Resources

About