Compositional pattern producing networks: A novel abstraction of development

By Kenneth O. Stanley. 2007

862 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Generative Models

Citations in the corpus

Perceiver IO: A General Architecture for Structured Inputs & Outputs (2021, methodology): “1), we use a 2D Fourier feature positional encoding (Vaswani et al., 2017; Stanley, 2007; Mildenhall et al., 2020; Tancik et al., 2020) using a sine and cosine bands with frequencies spaced linearly from a minimum frequency to a maximum frequency.”
Perceiver: General Perception with Iterative Attention (2021, methodology): “Here, we use a strategy that has recently gained renewed prominence, both in language and in vision: Fourier feature position encodings (Stanley, 2007; Vaswani et al., 2017; Parmar et al., 2018; Tancik et al., 2020; Mildenhall et al., 2020).”
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images (2014, background): “Indirectly encoded images tend to be regular because elements in the genome can affect multiple parts of the image [28].”
Hierarchical Representations for Efficient Architecture Search (2017, background): “The beneﬁts of indirect encoding schemes, such as multi-scale representations, have historically been discussed in Gruau et al. (1994); Kitano (1990); Stanley (2007); Stanley et al. (2009).”
A genetic programming approach to designing convolutional neural network architectures (2017, background): “e DPPN is a dierentiable version of the compositional paern-producing networks (CPPNs) [30].”
Evolving Deep Neural Networks (2017, background): “[10] evolved a CPPN (compositional paern-producing network [42]) to output the weights of an auto-encoder neural network.”
Convolution by Evolution: Differentiable Pattern Producing Networks (2016, methodology): “Compositional Pattern Producing Networks (CPPN) [25] were a major advance in evolutionary computation because they permitted evolution to eﬃciently optimize a model incrementally starting from a small number of parameters.”
Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data (2019, background): “More broadly, this type of parameterization is intensively explored named compositional pattern producing networks [73].”
An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution (2018, background): “Compositional Pattern Producing Networks (CPPNs) [31] implement functions from coordinates in arbitrarily many dimensions to one or more output values.”
Diffusion-based neuromodulation can eliminate catastrophic forgetting in simple neural networks (2017, background): “A prior paper on neuromodulation has already shown that a CPPN can specify the learning rule for connections (i.e. parameters for Hebbian or neuromodulation learning) based on their geometric positions within an ANN [80].”
AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search (2019, background): “Evolutionary Algorithm (EA) : Evolutionary algorithms represent each neural network as a string of genes and search the architecture space by mutation and recombinations [28, 38, 37, 13, 7, 20, 33, 43, 27, 39, 32].”
A novel generative encoding for exploiting neural network sensor and output geometry (2007, background): “Compositional Pattern Producing Networks (CPPNs) are a novel abstraction of development that can represent sophisticated repeating patterns in Cartesian space [15, 16].”
Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations (2019, background): “Compositional Pattern Producing Networks [54, 55] learn functions that map 2D image coordinates to color.”
Ieee Transactions on Evolutionary Computation on the Performance of Indirect Encoding across the Continuum of Regularity (neutral): “This figure is adapted from Stanley [16].”
Image Classification Using Generative Neuro Evolution for Deep Learning (2015, methodology): “HyperNEAT trains an indirect encoding, called a compositional pattern producing network (CPPN; [22]) that represents artificial neural network (ANN) connection weights.”