Evolving Neural Control Systems

By N. Saravanan, David B. Fogel. 1995

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

A Taxonomy for Artificial Embryogeny (2003, methodology): “For example, pole balancing has been used as a reinforcement learning benchmark for over 30 years [4, 7, 32, 35, 56, 58, 67, 76, 81–84], making it convenient for comparison with other methods.”
Efficient Reinforcement Learning Through Evolving Neural Network Topologies (2002, methodology): “In this paper, we focus on the pole balancing domain because it has been used as a reinforcement learning benchmark for over 30 years (Anderson 1989; Barto et al. 1983; Gomez and Miikkulainen 1999; Gruau et al. 1996; Michie and Chambers 1968; Moriarty and Miikkulainen 1996; Saravanan and Fogel 1995;”
Solving Non-Markovian Control Tasks with Neuro-Evolution (1999, methodology): “Other approaches that have applied shaping to the easier double pole task (with velocities) have incremented the short pole by only 1%[Wieland, 1991; Saravanan and Fogel, 1995].”
Neuroevolution strategies for episodic reinforcement learning (2009, methodology): “If after 10 evaluations no successful control strategy was found, a trial was regarded as a failure In such a case, the number of evaluations was set to 10 when computing the”
Spectrum-Diverse Neuroevolution With Unified Neural Models (2017, background): “Many works investigated the evolution of ﬁxed-topology neural networks [3], [4], [5], [6].”
Evolving artificial neural networks (1999, methodology): “Real numbers have been proposed to represent connection weights directly, i.e., one real number per connection weight [27], [29], [30], [48], [63]–[65], [74], [95], [96], [102], [110], [111], [117], [118].”
Evolving plastic neural networks with novelty search (2010, unknown): “…that evolved neural networks, that is, neuroevolution methods, which traditionally evolve either fixed-topology networks (Gomez & Miikkulainen, 1999; Saravanan & Fogel, 1995) or arbitrary random-topology networks (Angeline, Saunders, & Pollack, 1994; Gruau, Whitley, & Pyeatt, 1996; Yao, 1999).”
Accelerated Neural Evolution through Cooperatively Coevolved Synapses (2008, methodology): “Programming (EP; Saravanan and Fogel, 1995) is a general mutation-based evolutionary method that can be used to search the space of neural networks.”
On the deleterious effects of a priori objectives on evolution and representation (2011, background): “This insight is sobering because the vast majority of empirical experiments in EC assess their results based on benchmarks with a priori objectives [2, 8, 13, 20, 24, 27] and theory often focuses on whether targets can be reached [3, 17].”
Compositional Pattern Producing Networks : A Novel Abstraction of Development (2007, background): “neuroevolutionmethods, evolved either fixed-topology networks [30, 66], or arbitrary random-topology networks [5, 32, 82], NEAT is the first to begin evolution with a population of small, simple networks and complexifythe network topology over generations, leading to increasingly sophisticated behav”
A Hypercube-Based Encoding for Evolving Large-Scale Neural Networks (2009, background): “, neuroevolution methods), which traditionally evolve either fixed-topology networks [18, 45] or arbitrary random-topology networks [3, 22, 60].”
Evolving Artiicial Neural Networks (2007, methodology): “For example, the evolutionary approach has been used to train recurrent ANNs [41, 60, 126, 117, 65, 100, 102, 103, 106, 127, 128], higher-order ANNs [52, 53], and fuzzy ANNs [76, 77, 129, 130].”
Robust non-linear control through neuroevolution (2003, methodology): “Evolutionary Programming (EP; Saravanan and Fogel 1995) is a general mutation-based evolutionary method that can be used to search the space of neural networks.”
Sequential Constant Size Compressors for Reinforcement Learning (2011, methodology): “CoSyNE was shown [7] to be highly efficient, besting many other methods including both single-agent methods such as Adaptive Heuristic Critic [2], Policy Gradient RL [23], and evolutionary methods like SANE, ESP, NEAT [21], Evolutionary Programming [ 16 ], CMA-ES [9], and Cellular Encoding [8].”
Neuroevolution for reinforcement learning using evolution strategies (2003, methodology): “Saravanan and Fogel (1995) applied evolutionary programming (EP) to parameter optimization of feedforward NNs for the double pole balancing task.”
Compositional pattern producing networks: A novel abstraction of development (2007, background): “neuroevolution methods, evolved either fixed-topology networks [63, 64], or arbitrary random-topology networks [21, 56, 65], NEAT is the first to begin evolution with a population of small, simple networks and complexify the network topology over generations, leading to increasingly sophisticated be”
Evolving Neural Networks through Augmenting Topologies (2002, methodology): “hidden units like the fixed topology methods in this task (Saravanan and Fogel, 1995; Wieland, 1991).”