Evolutionsstrategie : Optimierung technischer Systeme nach Prinzipien der biologischen Evolution

By I. Rechenberg. 1973

How researchers see this paper

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

Citations in the corpus

Deep learning in neural networks: An overview (2014, background): “adients of the total reward with respect to policies (NN weights) are estimated (and then exploited) through repeated NN evaluations. RL NNs can also be evolved through Evolutionary Algorithms (EAs) (Rechenberg, 1971; Schwefel, 1974; Holland, 1975; Fogel et al., 1966; Goldberg, 1989) in a series of ”
World Models (2018, background): “This choice allows us to explore more unconventional ways to train C – for example, even us-ing evolution strategies (ES) (Rechenberg, 1973; Schwefel, 1977) to tackle more challenging RL tasks where the credit assignment problem is difﬁcult.”
Deep Networks with Internal Selective Attention through Feedback Connections (2014, background): “Future research will also consider more complex actions that spatially focus on (or alter) parts of observed images.”
PowerPlay: Training an Increasingly General Problem Solver by Continually Searching for the Simplest Still Unsolvable Problem (2011, methodology): “cable, see Section 3.1.2). Use the information stored so far to adapt the parameters θ of the BBOA, e.g., by gradient-based search [60, 49], or according to the principles of evolutionary computation [24, 10, 60]. end for 5 Outgrowing Trivial Tasks - Compressing Previous Solutions What prevents POWE”
Evolution Strategies as a Scalable Alternative to Reinforcement Learning (2017, methodology): “Evolution Strategies (ES) is a class of black box optimization algorithms [Rechenberg and Eigen, 1973, Schwefel, 1977] that are heuristic search procedures inspired by natural evolution: At every iteration (“generation”), a population of parameter vectors (“genotypes”) is perturbed (“mutated”) and…”
The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities (2018, neutral): “In particular, the University of Freiburg team employed a simple version of a decades-old EA called evolution strategies [114].”
Parameter Space Noise for Exploration (2017, background): “Our work is also closely related to evolution strategies (ES) [39, 40], and especially neural evolution strategies (NES) [41, 8, 42–45].”
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.”
Optimal Ordered Problem Solver (2002, methodology): “Unlike logicbased program synthesizers (Green, 1969, Waldinger and Lee, 1969, Deville and Lau, 1994), program evolvers use biology-inspired concepts of Evolutionary Computation (Rechenberg, 1971, Schwefel, 1974) or Genetic Algorithms (Holland, 1975) to evolve better and better computer programs.”
Flipout: Efficient Pseudo-Independent Weight Perturbations on Mini-Batches (2018, positive): “ES (Rechenberg & Eigen, 1973) is a family of black box optimization algorithms which use weight perturbations to search for model parameters.”
Evolved Policy Gradients (2018, methodology): “Due to the lack of easily exploitable structure in this optimization problem, we turn to evolution strategies (ES) (35; 46; 16; 37) as a blackbox optimizer.”
Evolution strategies – A comprehensive introduction (2002, unknown): “Figure 10 displays P s and the normalized progress rate ϕ * calculated for the sphere model (16) in the asymptotic limit case N → ∞ (Rechenberg, 1973; Beyer, 2001b ).”
Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation (2001, methodology): “The ES ( Rechenberg, 1973; Schwefel, 1977) approach is somewhat difierent from the GA approach presented in the previous subsection.”
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].”
A Survey of Optimization by Building and Using Probabilistic Models (2000, methodology): “Similar approaches are used in self-adaptive evolution strategies (ES) [37].”
Fitness landscape analysis and memetic algorithms for the quadratic assignment problem (2000, unknown): “programming [66], evolution strategies[67], genetic algo-”
Towards industrial ecology: sustainable development as a concept of ecological modernization (2000, unknown): “…Ayres & Simonis, 1994; Ayres & Ayres, 1996), as well as design for environment (Kreibich et al., 1991; Stahel, 1991, 1992; Paton, 1994), bionics (Rechenberg, 1973; Gleich, 1998), ecoeffectiveness (Braungart & McDonough, 1999), constructive technology assessment (Rip et al., 1995), ecological…”
Self-Organizing Pedestrian Movement (2001, methodology): “Their configuration and shape can be varied systematically, for example, by means of evolutionary algorithms (Baeck, 1996; Bolay, 1998; Rechenberg, 1973) (see figure 9), and evaluated on the basis of particular mathematical performance measures (Helbing, 1997; Helbing and Molnär, 1997).”