How deeply does Deep Learning with R, Third Edition cover fully connected neural network?

Defines all-to-all adjacent-layer connections, contrasts local convolution, and demonstrates practical classification and regression configurations.

Fully connected neural network

As explained in Deep Learning with R, Third Edition

A dense neural network is a neural network architecture in which units in one layer are connected to all units in the adjacent layer.

A fully connected neural network, also called a densely connected network, stacks Dense layers whose units connect to every unit in the adjacent layer. It processes vector data without assuming a particular feature structure, making it a practical baseline for tabular and structured data. section 20.1.6

Table 20.1 Model architectures for different data types

Input	Output	Model
Vector data	Class probability, regression value	Densely connected network
Timeseries data	Class probability, regression value	RNN, Transformer
Images	Class probability, regression value	Convnet
Text	Class probability, regression value	Transformer
Text, images	Text	Transformer
Text, images	Images	VAE, diffusion model

Use the architecture table to match fully connected networks with vector inputs and classification or regression outputs, then compare them with modality-specific alternatives. section 20.1.6

Choose the output layer and loss according to the prediction task:

Binary classification: one sigmoid unit with binary crossentropy. section 20.1.6
Single-label categorical classification: one softmax unit per class with categorical crossentropy. section 20.1.6
Multilabel classification: one sigmoid unit per class with binary crossentropy. section 20.1.6
Regression: use an output layer without an activation and mean squared error. section 13.2.3

A simple forecasting baseline

For Jena temperature forecasting, reshape each input sequence into a vector, pass it through Dense(16, activation = "relu") and Dense(1), then compile with Adam, MSE, and MAE. Save the best validation model and evaluate it on the test set. This inexpensive model provides a baseline before trying recurrent networks. section 13.2.3

Dense connectivity can represent relationships between any pair of input features, unlike convolutional layers that emphasize local relationships. That flexibility is useful when features have no known spatial or sequential structure, but it can be a poor choice when the data demands a stronger architecture prior. section 20.1.6 section 5.3.2

How Deep Learning with R, Third Edition covers it

Cover of Deep Learning with R, Third Edition

liveBook manning.com

Read Deep Learning with R, Third Edition on liveBook Get the book at manning.com

Also explained in

liveBook manning.com

Grokking Machine LearningA neural-network layer in which each unit receives input from all units in the preceding layer.

Compare how 2 books explain fully connected neural network →

Related concepts

N-BEATSused by · N-BEATS is a neural forecasting model organized as stacks of blocks for modeling time series without requiring explicit series decomposition or input transformations.
deep learningkind of · Deep learning is a machine-learning approach that uses neural networks with multiple computational layers to learn representations from data.
linear layerhas part · A linear layer is a neural-network layer that maps input features to outputs by a learned affine transformation, typically connecting each output to every input feature.
recurrent neural networkcontrasts with · A recurrent neural network is a neural network that processes sequences by updating an internal state across successive inputs, allowing earlier inputs to influence later computations.
feedforward networkequivalent to · A feedforward network is a neural network or subnetwork in which information flows from inputs toward outputs through successive computations without recurrent connections or cycles.
sequencecontrasts with · A sequence is an ordered collection of elements whose positions determine their order of traversal or processing.

This explainer was generated by Manning's knowledge graph from the text of Deep Learning with R, Third Edition. The section links point to the parts of the book it draws on.