Neuroevolution for reinforcement learning using evolution strategies

By Christian Igel. 2003

181 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

Deep learning in neural networks: An overview (2014, methodology): “Related methods include probability distributionbased EAs (Baluja, 1994; Larraanaga & Lozano, 2001; Sałustowicz & Schmidhuber, 1997; Saravanan & Fogel, 1995), Covariance Matrix Estimation Evolution Strategies (CMA-ES) (Hansen, Müller, & Koumoutsakos, 2003; Hansen & Ostermeier, 2001; HeidrichMeisner ”
Evolving Deep Neural Networks (2017, methodology): “CMA-ES [24], a technique for continuous optimization, works well on optimizing the”
On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models (2015, methodology): “1, except that C is now an FNN or RNN trained by evolutionary algorithms [200, 255, 105, 56, 68] applied to NNs [165, 321, 180, 259, 72, 90, 89, 110, 94], or by policy gradient methods [314, 315, 316, 274, 18, 1, 63, 128, 313, 210, 192, 191, 256, 85, 312, 190, 82, 93][245, Sec.”
Neuroevolution (2010, unknown): “For instance, intelligent mutation techniques such as those employed in evolutionary strategies are effective because the weights often have suitable correlations [9].”
Trends and Applications in Constructive Approximation Evolutionary Optimization of Neural Systems: the Use of Strategy Adaptation (unknown): “We sketch applications of these methods to the design of different types of neural systems: topology optimization of multilayer NNs for face detection [49], weight optimization of recurrent networks for solving reinforcement learning tasks [21], and model selection for support vector machines [14].”
Multifactorial Evolutionary Algorithm With Online Transfer Parameter Estimation: MFEA-II (2020, background): “In literature, the double-pole balancing problem (or inverted pendulum system) has been a standard benchmark for artiﬁcial learning systems [56, 66–69].”
Autoencoder-augmented neuroevolution for visual doom playing (2017, background): “Neuroevolution has long been known to be an effective reinforcement learning method, in particular when training comparatively small networks [11], [7].”
Accelerated Neural Evolution through Cooperatively Coevolved Synapses (2008, unknown): “, 2007), CMA-ES (Igel, 2003), EP (Saravanan and Fogel, 1995), and CE2 (Gruau et al.”
Evolutionary Computation for Reinforcement Learning (2012, methodology): “…context, this internal state can record aspects of the agent's observation history , which can help it cope with partial observability (Wieland, 1991; Gomez and Miikkulainen, 1999; Moriarty et al, 1999; Stanley and Miikkulainen, 2002; Igel, 2003; Gomez and Schmidhuber, 2005a,b).”
Neuroevolution strategies for episodic reinforcement learning (2009, methodology): “In more biological terms (e.g., see [10]), the CMA-NeuroES implements trial-and-error learning based on extrinsic”
Policy Optimization by Genetic Distillation (2017, methodology): “CMA-ES has been used to learn neural network policies for reinforcement learning (CMA-NeuroES, Heidrich-Meisner & Igel (2009); Igel (2003)).”
Neuroevolution: from architectures to learning (2008, background): “Recent benchmark experiments with Evolution Strategies [62], which use a floating-point representation of the synaptic weights, have reported excellent performance with direct encoding of a small, fixed architecture [38].”
Cartesian genetic programming encoded artificial neural networks: a comparison using three benchmarks (2013, unknown): “CMA-ES [12] 895 50”