Bayesian neural network
A Bayesian neural network is a neural network that represents uncertain parameters or functions with probability distributions to account for uncertainty in its predictions.
Grokking Bayes
A Bayesian neural network is a neural network whose uncertain parameters or predictions are represented with probability distributions.
- How it is used
- Combines deep-learning function approximation with Bayesian uncertainty quantification for predictions.
- How deep it goes
- Explains priors, likelihoods, posterior sampling, credible intervals, and inference trade-offs, but does not derive posterior calculations.
- Examples
- Uses noisy regression, two hidden layers, posterior samples, predictive means, and widening 95% credible intervals.
- What you can do after
- Build, train, and perform inference with Bayesian neural networks using MCMC, variational inference, or Monte Carlo dropout.
Timeless Algorithms, The Foundational Ideas
A Bayesian neural network is a neural network whose weights and biases are represented by probability distributions, allowing predictions to account for parameter uncertainty.
- How it is used
- Extends fixed-parameter neural networks to represent uncertainty about their weights and biases.
- How deep it goes
- Defines parameter distributions and Bayesian updating, but stops short of detailing inference algorithms.
- Examples
- Uses a pedestrian-presence probability of 0.68 to distinguish class ambiguity from uncertainty about parameters.
- What you can do after
- Understand why a fixed-parameter prediction does not express uncertainty about parameter values and how data update a prior.
Read how Timeless Algorithms, The Foundational Ideas explains bayesian neural network →
What to know first
- neural networkA neural network is a computational model composed of interconnected, parameterized units or transformations that learn to map inputs to outputs.
- Bayesian inferenceBayesian inference updates probability distributions over hypotheses or unknown quantities by combining prior information with observed evidence and a likelihood model.
- prior probabilityA prior probability or distribution is a probability or probability distribution assigned to an event, hypothesis, or unknown quantity before relevant evidence is incorporated.
Related concepts
- neural networkkind of · A neural network is a computational model composed of interconnected, parameterized units or transformations that learn to map inputs to outputs.
- prior probabilityuses · A prior probability or distribution is a probability or probability distribution assigned to an event, hypothesis, or unknown quantity before relevant evidence is incorporated.
- variational inferenceused by · Variational inference is an approximate inference technique that optimizes a tractable family of probability distributions to approximate a target distribution, often a posterior.
- deep learninguses · Deep learning is a machine-learning approach that uses neural networks with multiple computational layers to learn representations from data.
- markov chain monte carloused by · Markov chain Monte Carlo is a family of sampling algorithms that uses a Markov chain to generate samples from a target probability distribution.
- networkkind of · A network is a set of interconnected nodes and links that represents relationships or enables communication and exchange among the nodes.
Questions about bayesian neural network
What is bayesian neural network?
A Bayesian neural network is a neural network that represents uncertain parameters or functions with probability distributions to account for uncertainty in its predictions.
Which Manning books explain bayesian neural network?
Grokking Bayes and Timeless Algorithms, The Foundational Ideas.
What should I know before learning bayesian neural network?
Manning books that teach bayesian neural network usually introduce neural network, Bayesian inference and prior probability first.
What is bayesian neural network a kind of?
Bayesian neural network is a kind of neural network. A neural network is a computational model composed of interconnected, parameterized units or transformations that learn to map inputs to outputs.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

