Adding Learning to the Cellular Development of Neural Networks: Evolution and the Baldwin Effect

By Frédéric Gruau, L. D. Whitley. 1993

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

Evolving Deep Neural Networks (2017, methodology): “Further, techniques such as Cellular Encoding [18] and NEAT [43] have been developed to evolve the topology of the neural network, which is particularly effective in determining the required recurrence.”
Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks (2017, background): “One early study (Gruau and Whitley, 1993) encoded plasticity and development with a grammar tree, and compared different learning rules on a simple static task (parity and symmetry), demonstrating that learning provided an evolutionary advantage in a static scenario.”
Symbiotic Evolution of Neural Networks in Sequential Decision Tasks (1997, background): “Several researchers have investigated the use of local learning after each action in an evolutionary algorithm (Grefenstette 1991; Gruau and Whitley 1993; Littman 1995; Nol and Parisi 1995).”
Learning to Adapt to Changing Environments in Evolving Neural Networks (1996, result): “Similar results have been reported for more complex models that do not have some of the simplifications of Hinton and Nowlan's model (Ackley and Littman, 1991; Gruau and Whitley, 1993; Nolfi et al., 1994a; cf. also Parisi and Nolfi, 1995).”
Neuroevolution (2010, background): “Indirect encoding, in contrast, specifies a process through which the network is constructed, such as cell division or generation through a grammar [5, 8, 17, 21].”
On the Relationships between Generative Encodings, Regularity, and Learning Abilities when Evolving Plastic Artificial Neural Networks (2013, background): “Several papers report experiments in which neural networks with synaptic plasticity are evolved [2,15,18–24].”
Evolving Plastic Neural Networks for Online Learning: Review and Future Directions (2012, background): “Gruau [13] developed a grammar tree system where the values of weights as determined by plasticity rules could be passed onto new connections created by recursive expansion of the grammar tree during an individuals life.”
Evolutionary Function Approximation for Reinforcement Learning (2006, methodology): “This biologically intuitive combination has been applied to computational systems in the past [1, 5, 8, 10, 12, 18] but never, to my knowledge, to aid the”
Evolutionary Computation for Reinforcement Learning (2012, methodology): “Similarly, cellular encodings (Gruau and Whitley, 1993; Gruau, 1994) evolve graph grammars for generating modular neural networks composed of simpler subnetworks.”
Empirical Investigation of the Benefits of Partial Lamarckianism (1997, background): “Hinton and Nolan showed that, without learning, the genetic algorithm fails; however, when hybridized with the random search, the GA is capable of finding the optimum.”
Cost-Sensitive Classification: Empirical Evaluation of a Hybrid Genetic Decision Tree Induction Algorithm (1994, background): “This phenomenon (and other phenomena that result from this form of hybrid) is known as the Baldwin effect (Baldwin, 1896; Morgan, 1896; Waddington, 1942; Maynard Smith, 1987; Hinton & Nowlan, 1987; Ackley & Littman, 1991; Whitley & Gruau, 1993; Whitley et al., 1994; Anderson, in press).”
Hybrid Genetic Algorithms: A Review (2006, methodology): “Gruau and Whitley [11] showed how local search can change the landscape of fitness function into flat landscapes around the basin of attraction.”