Skip to content

Feedforward Network (FFN)

Definition
AI-generated

A feedforward network (FFN) is a computation graph without cycles: each layer maps its input through weights and a (usually nonlinear) activation to produce the next layer’s activations.

Topics
LLM and Agents Machine learning concepts
Synonyms
Feedforward neural network FFN / FFNN / MLP (informal overlaps)
Related
Deep Learning Genome-Wide Association Study (GWAS) Multilayer Perceptron (MLP)

Why it matters in GWAS

FFN layers stack with attention in sequence models for DNA and protein; capacity and depth interact with regularization when predicting variant effects or cell states alongside classical GWAS.

Example usage

"We doubled the transformer FFN hidden ratio from 4× to 8× but saw no gain on the held-out regulatory validation set."

Deep Learning Genome-Wide Association Study (GWAS) Multilayer Perceptron (MLP) Neural Network Self-Attention 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

  • Vaswani A, et al. (2017). Attention is all you need. NeurIPS.
  • Hornik K, Stinchcombe M, White H. (1989). Multilayer feedforward networks are universal approximators. Neural Netw.
← FASTQ Fertilization →

Last updated (UTC · Git history)