Skip to content

Neural Network

Definition
AI-generated

A neural network is a parametric function built from layers of connected units (nodes) with adjustable weights; training uses labeled or unlabeled data and a loss function, typically optimized by gradient-based methods so the network approximates a mapping from inputs to outputs.

Topics
LLM and Agents
Synonyms
Artificial neural network (ANN)
Related
Backpropagation Convolutional Neural Network (CNN) Deep Learning

Why it matters in GWAS

Neural models appear in polygenic prediction, variant effect and splicing predictors, single-cell embedding and batch correction, and generative simulation of molecular or phenotypic data. Architecture choice, leakage from relatedness or population structure, and limited portability across ancestries affect whether gains over simpler linear or mixed models hold in genetic studies.

Example usage

"The phenotype imputation step compared elastic net baselines against a shallow neural network trained on the UK Biobank field dictionary."

Backpropagation Convolutional Neural Network (CNN) Deep Learning Deep Generative Model Generalized Linear Mixed Model (GLMM) Graph Neural Network (GNN) Machine Learning Multilayer Perceptron (MLP) Recurrent Neural Network (RNN) Transformer
Related-term network

Click any node to open that term page.

Node order. The current term is first. Next come every page linked from Related terms, in the same top-to-bottom order as that list. Then new terms appear in two-hop layers: for each of those direct neighbors (in that same order), their related terms are appended if not already listed. A third pass does the same for two-hop nodes. The walk stops when a fixed maximum number of nodes is reached.

References

  • Rumelhart DE, Hinton GE, Williams RJ. (1986). Learning representations by back-propagating errors. Nature.
← Net Reclassification Improvement (NRI) Neutral theory of molecular evolution →

Last updated (UTC · Git history)