Neuroevolution: from architectures to learning

By Dario Floreano, Peter Dürr, Claudio Mattiussi. 2008

761 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Evolutionary Computation

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

In Evolutionary Computation

Citations in the corpus

BANANAS: Bayesian Optimization with Neural Architectures for Neural Architecture Search (2019, background): “r or superior to several state-of-the-art NAS algorithms on two dierent search spaces. 2 Related Work Neural architecture search. Neural architecture search has been studied since at least the 1990s [7, 18, 42], but the eld was revitalized in 2017 when NASNet gained signicant attention [54]. NASNet ”
A Survey on Neural Architecture Search (2019, neutral): “Shortly thereafter, Real et al. (2017) showed that similar results could also be achieved by the hitherto well studied approaches in neuroevolution (Floreano et al., 2008).”
Evolutionary-Neural Hybrid Agents for Architecture Search (2018, neutral): “Prior works have used evolutionary methods to evolve neural networks [8, 16, 34, 38, 45].”
MaskConnect: Connectivity Learning by Gradient Descent (2018, positive): “Evolutionary search has been proposed as an interesting framework to learn both the structure as well as the connections in a neural network [16,17,18,19,20,21,22,23,24].”
Neural Architecture Search with Bayesian Optimisation and Optimal Transport (2018, background): “#18 conv3, 256, (32768) [8] #19 conv3, 256, (32768) [8]”
Connectivity Learning in Multi-Branch Networks (2017, methodology): “itecture or connectivity learning. Evolutionary search has been proposed as an interesting framework to learn both the structure as well as the connections in a neural network (Wierstra et al., 2005; Floreano et al., 2008; Real et al., 2017). Architecture search has also been recently formulated as ”
SMASH: One-Shot Model Architecture Search through HyperNetworks (2017, background): “Evolutionary techniques [10, 37, 38, 40] offer a ﬂexible approach for discovering variegated models from trivial initial conditions, but often struggle to scale to deep neural nets where the search space is vast, even with enormous compute power [26].”
DeepArchitect: Automatically Designing and Training Deep Architectures (2017, neutral): “[9], and Real et al .”
Evolving Deep Neural Networks (2017, methodology): “Neuroevolution techniques have been applied successfully to sequential decision tasks for three decades [11, 31, 36, 50].”
Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks (2017, background): “Just as neuromodulation is one particular plasticity mechanism that enables effective reward learning, other principles may beneﬁt particular types of learning behavior, as outlined in the review by (Floreano et al., 2008) and in a recent book by Downing (2015).”
A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, background): “Neuroevolution refers to using evolutionary algorithms to train artiﬁcial neural networks [70].”
Neuroevolution (2010, background): “The idea is that topology can have a large effect on function, and evolving appropriate topologies can achieve good performance faster than evolving weights only [2, 5, 18, 21].”
Evolution of Plastic Learning in Spiking Networks via Memristive Connections (2012, methodology): “A survey of various methods for evolving both weights and architectures in neural networks is presented in [20].”
Genetic Representation and Evolvability of Modular Neural Controllers (2010, methodology): “In this context, Artificial Neural Networks (ANNs) are often used as control architectures because they can approximate arbitrary mappings from sensory inputs to actuator outputs [4], [5].”
Artificial Evolution of Plastic Neural Networks: A Few Key Concepts (2014, background): “A comparatively few papers proposed to combine the artiﬁcial evolution of neural networks with synaptic plasticity to evolve Jean-Baptiste Mouret and Paul Tonelli ISIR, Universit´e Pierre et Marie Curie-Paris 6, CNRS UMR 7222, Paris, France e-mail: {mouret,tonelli}@isir.upmc.fr”
Towards evolving spiking networks with memristive synapses (2011, methodology): “A survey of various methods for evolving both weights and architectures in neural networks is presented in [10].”
On the Relationships between Generative Encodings, Regularity, and Learning Abilities when Evolving Plastic Artificial Neural Networks (2013, methodology): “Several papers report experiments in which neural networks with synaptic plasticity are evolved [2,15,18–24].”