Evolving Neural Networks through Augmenting Topologies
- Kenneth O. Stanley, R. Miikkulainen
- BiologyEvolutionary Computation
- 1 June 2002
A method is presented, NeuroEvolution of Augmenting Topologies (NEAT), which outperforms the best fixed-topology method on a challenging benchmark reinforcement learning task and shows how it is possible for evolution to both optimize and complexify solutions simultaneously.
Abandoning Objectives: Evolution Through the Search for Novelty Alone
- J. Lehman, Kenneth O. Stanley
- PsychologyEvolutionary Computation
- 1 June 2011
In the maze navigation and biped walking tasks in this paper, novelty search significantly outperforms objective-based search, suggesting the strange conclusion that some problems are best solved by methods that ignore the objective.
Compositional pattern producing networks: A novel abstraction of development
- Kenneth O. Stanley
- Computer ScienceGenetic Programming and Evolvable Machines
- 1 June 2007
Results produced with CPPNs through interactive evolution of two-dimensional images show that such an encoding can nevertheless produce structural motifs often attributed to more conventional developmental abstractions, suggesting that local interaction may not be essential to the desirable properties of natural encoding in the way that is usually assumed.
Exploiting Open-Endedness to Solve Problems Through the Search for Novelty
- J. Lehman, Kenneth O. Stanley
- Computer ScienceIEEE Symposium on Artificial Life
- 2008
Decoupling the idea of open-ended search from only artificial life worlds, the raw search for novelty can be applied to real world problems and significantly outperforms objective-based search in the deceptive maze navigation task.
A Hypercube-Based Encoding for Evolving Large-Scale Neural Networks
- Kenneth O. Stanley, David B. D'Ambrosio, J. Gauci
- Biology, Computer ScienceArtificial Life
- 1 April 2009
The main conclusion is that the ability to explore the space of regular connectivity patterns opens up a new class of complex high-dimensional tasks to neuroevolution.
Evolving a diversity of virtual creatures through novelty search and local competition
- J. Lehman, Kenneth O. Stanley
- Computer ScienceAnnual Conference on Genetic and Evolutionary…
- 12 July 2011
The results in an experiment evolving locomoting virtual creatures show that novelty search with local competition discovers more functional morphological diversity within a single run than models with global competition, which are more predisposed to converge.
Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning
- F. Such, Vashisht Madhavan, Edoardo Conti, J. Lehman, Kenneth O. Stanley, J. Clune
- Computer ScienceArXiv
- 18 December 2017
It is shown that combining DNNs with novelty search, which was designed to encourage exploration on tasks with deceptive or sparse reward functions, can solve a high-dimensional problem on which reward-maximizing algorithms fail, and expands the sense of the scale at which GAs can operate.
Search-Based Procedural Content Generation: A Taxonomy and Survey
- J. Togelius, Georgios N. Yannakakis, Kenneth O. Stanley, C. Browne
- Computer ScienceIEEE Transactions on Computational Intelligence…
- 29 April 2011
This article contains a survey of all published papers known to the authors in which game content is generated through search or optimisation, and ends with an overview of important open research problems.
Quality Diversity: A New Frontier for Evolutionary Computation
- Justin K. Pugh, L. Soros, Kenneth O. Stanley
- Computer ScienceFrontiers in Robotics and AI
- 12 July 2016
A new approach is investigated that hybridizes multiple views of behaviors (called behavior characterizations) in the same run, which succeeds in overcoming some of the challenges associated with searching for QD with respect to a behavior characterization that is not necessarily sufficient for generating both quality and diversity at the same time.
Efficient evolution of neural networks through complexification
- Kenneth O. Stanley, R. Miikkulainen
- Computer Science
- 2004
This dissertation presents the NeuroEvolution of Augmenting Topologies (NEAT) method, which makes search for complex solutions feasible and is first shown faster than traditional approaches on a challenging reinforcement learning benchmark task, and used to successfully discover complex behavior in three challenging domains.
...
...