Clinical applications of artificial neural networks: Evolving artificial neural networks

By V. W. Porto, David B. Fogel. 2001

1,187 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Medical & Health AI

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 Medical & Health AI

Citations in the corpus

Primer: Searching for Efficient Transformers for Language Modeling (2021, methodology): “Given a search space of TensorFlow programs, we use evolution [11, 12, 13, 14, 15, 16, 17] to search for models that achieve as low of a validation loss as possible given a fixed amount of training compute.”
Regularized Evolution for Image Classifier Architecture Search (2018, background): “Review papers provide informative surveys of earlier (Yao 1999; Floreano, Dürr, and Mattiussi 2008) and more recent”
Hierarchical Representations for Efficient Architecture Search (2017, methodology): “The evolution of neural architectures was studied as a sub-task of neuroevolution (Holland, 1975; Miller et al., 1989; Yao, 1999; Stanley & Miikkulainen, 2002; Floreano et al., 2008), where the topology of a neural network is simultaneously evolved along with its weights and hyperparameters.”
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): “…the ﬁeld of evolutionary computation (Holland, 1975; Eiben and Smith, 2015) includes areas such as evolutionary robotics (Harvey et al., 1997; Nolﬁ and Floreano, 2000) and neuroevolution (Yao, 1999) to investigate the autonomous design of complex, bio-inspired adaptive physical and neural systems.”
Incremental Learning in Biological and Machine Learning Systems (2002, background): “Several reviews show that evolutionary artificial neural networks have been investigated and applied successfully in recent years.(104,125,142,146) According to Yao(146) evolution in artificial neural networks can include adaption of connection weights (an alternative to standard network training, s”
The Importance of Neurophysiological Constraints for Modelling the Emergence of Modularity (neutral): “The idea of applying simulated evolution by natural selection to neural networks is now well established (e.g., Yao, 1999; Cantû-Paz & Kamath, 2005).”
Online Evolution of Deep Convolutional Network for Vision-Based Reinforcement Learning (2014, background): “neuroevolution; [26]), but instead of accessing the observations directly, it receives feature vectors of much lower dimensionality provided by the unsupervised compressor, which is also evolved.”
Artificial Intelligence techniques: An introduction to their use for modelling environmental systems (2008, background): “They are highly interconnected by links (synapses) with weights [170] [133] [5].”
The road to everywhere : Evolution, complexity and progress in natural and artificial systems (2008, background): “An impressive survey of the application of evolutionary techniques to neural networks can be found in Yao [198] (who also favour mutation-only evolution).”
A New Adaptive Merging and Growing Algorithm for Designing Artificial Neural Networks (2009, background): “A large number of ANN design algorithms have been proposed based on evolutionary approaches (see the review papers [38]–[40]).”
Supplementing evolutionary developmental systems with abstract models of neurogenesis (2007, background): “However, bi-directional signalling is extremely prevalent among cerebral regions, indicating more of a heterarchical than purely hierarchical organization, as shown in Figure 1.”
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”
Indirectly Encoding Neural Plasticity as a Pattern of Local Rules (2010, methodology): “Additionally, as noted by Yao [10], the right learning rule depends on the ANN architecture, which makes it difficult to design an optimal such rule a priori.”
Particle swarm optimization in electromagnetics (2004, background): “Realizing the importance of exploration early in the run, and the increasing importance of exploitation of maxima as the run progresses, Eberhart and Shi [1] suggested varying the inertial weight linearly from 0.9 to 0.4 over the course of the run.”
Evolving Soccer Keepaway Players Through Task Decomposition (2005, methodology): “The learning algorithm in all cases is neuroevolution, which has been shown to be effective in learning control policies and behavior strategies in similar domains (Schaffer et al., 1992; Gomez and Miikkulainen, 2001; Yao, 1999).”
Evolving Least Squares Support Vector Machines for Stock Market Trend Mining (2009, background): “In addition, a large number of input variables will increase the size of LSSVM and thus require more training data and longer training times in order to obtain a reasonable generalization ability [44].”