Symbiotic Evolution of Neural Networks in Sequential Decision Tasks

By David E. Moriarty. 1997

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, neutral): “Many RNN evolvers have been proposed (e.g., Cliff, Husbands, & Harvey, 1993; Juang, 2004; Miglino, Lund, & Nolfi, 1995; Miller, Todd, & Hedge, 1989; Moriarty, 1997; Nolfi, Floreano, Miglino, & Mondada, 1994; Pasemann, Steinmetz, & Dieckman, 1999; Sims, 1994; Whiteson, 2012; Wieland, 1991; Yamauchi…”
On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models (2015, methodology): “1 RL through Direct and Indirect Search in RNN Program Space It is possible to train small RNNs with a few 100 or 1000 weights using evolutionary algorithms [200, 255, 105, 56, 68] to search the space of NN weights [165, 307, 44, 321, 180, 259, 320, 164, 173, 69, 71, 187, 121, 313, 66, 270, 269, 305”
One Big Net For Everything (2018, background): “Without a teacher, reward-maximizing programs of an RNN must be learned through repeated trial and error, e.g., through artiﬁcial evolution [39, 107, 83, 40, 16, 18, 103, 14, 89, 88] [76, Sec.”
Evolving Neural Networks through Augmenting Topologies (2002, background): “Past studies have shown NE to be faster and more efficient than reinforcement learning methods such as Adaptive Heuristic Critic and Q-Learning on single pole balancing and robot arm control (Moriarty and Miikkulainen, 1996; Moriarty, 1997).”
Training Recurrent Networks by Evolino (2007, background): “This approach can quickly learn t o solve difficult reinforcement learning control tasks [5,6,22], including ones that require use of deep memory [7].”
Cooperative Coevolution: An Architecture for Evolving Coadapted Subcomponents (2000, methodology): “For example, in the SANE system (Moriarty 1997; Moriarty and Miikkulainen 1997) each individual in the population represents a single neuron by specifying which input and output units it connects to and the weights on each of its connections.”
Real-time neuroevolution in the NERO video game (2005, methodology): “Using this approach, several NE algorithms have been applied to board games [11], [12], [63], [64].”
Forming Neural Networks Through Efficient and Adaptive Coevolution (1997, result): “Other simulations produced similar plots Moriarty, 1997.”
Accelerated Neural Evolution through Cooperatively Coevolved Synapses (2008, methodology): “Symbiotic, Adaptive Neuro-Evolution (SANE; Moriarty, 1997) is a cooperative coevolutionary method that evolves two different populations simultaneously: a population of neurons and a population of network blueprints that specify how the neurons are combined to form complete networks.”
Evolution of neural networks using Cartesian Genetic Programming (2010, background): “Also blueprints that results in good neuron combination are promoted to the next generation and recombined to produce even better combinations [ 17 ].”
Developmental Plasticity in Cartesian Genetic Programming based Neural Networks (2016, background): “Method SinglePole DoublePole CNE 352 22100 SANE 302 12600 ESP 289 3800 NEAT 743 3600 CoSyNE 98 954 FCGPANN 21 77 PCGPANN 104 1169 Evolution (SANE)(Moriarty, 1997), Enforced SubPopulation (ESP) (Gomez and Miikkulainen, 1999), Neuro Evolution of Augmenting Topologies (NEAT) (Stanley and Miikkulainen,…”
Robust non-linear control through neuroevolution (2003, methodology): “SANE evolves good networks more quickly because the network sub-functions are allowed to evolve independently.”
Active Guidance for a Finless Rocket Using Neuroevolution (2003, methodology): “Enforced Subpopulations1 [5,6] is a neuroevolution method that extends the Symbiotic, Adaptive Neuroevolution algorithm (SANE; [7]).”
2-D Pole Balancing with Recurrent Evolutionary Networks (1998, methodology): “SANE has been shown e ective in a variety of domains, including robot arm control, constraint satisfaction, and in controlling chaos [6, 8, 9].”
Incremental Evolution of Complex General Behavior (1997, methodology): “The SANE approach has proven faster and more e cient than other reinforcement learning methods (Moriarty and Miikkulainen 1996a; Moriarty 1997).”