Skip to content

Autoencoder

Definition
AI-generated

An autoencoder is a neural network trained to map inputs through a low-dimensional bottleneck (encoder) and reconstruct them (decoder), minimizing reconstruction error; it learns a compressed representation useful for denoising, anomaly detection, or pretraining.

Topics
LLM and Agents Machine learning concepts
Synonyms
AE
Related
Decoder Deep Generative Model Disentangled Representation Learning

Why it matters in GWAS

Autoencoders reduce dimensionality of multi-omic or single-cell count matrices, correct batch effects in latent space, and serve as building blocks inside deep generative pipelines. Latent factors can correlate with ancestry or technical noise, so downstream association or TWAS-style analyses need orthogonal checks.

Example usage

"We trained a denoising autoencoder on scRNA-seq counts, then associated latent factors with cis-eQTL genotypes."

Decoder Deep Generative Model Disentangled Representation Learning Encoder Neural Network Principal Component Analysis (PCA) Single-Cell Sequencing Variational Autoencoder (VAE)
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

  • Hinton GE, Salakhutdinov RR. (2006). Reducing the dimensionality of data with neural networks. Science.
← Attention Mechanism Autosome →

Last updated (UTC · Git history)