How deeply does Outlier Detection in Python cover bidirectional generative adversarial network?

Explains encoder, generator, discriminator, alternating training, reconstruction, and scoring without deriving the adversarial objective.

Bidirectional generative adversarial network

As explained in Outlier Detection in Python

A bidirectional generative adversarial network is a GAN augmented with an encoder that maps observations back into the generator's latent representation.

A bidirectional generative adversarial network (BiGAN) augments a GAN with an encoder E, or inverse mapper, that maps observations into the generator's latent space. The generator G produces realistic data from latent input, while the discriminator D distinguishes real from generated data, often using both observations and latent representations. section 16.2.3

Use the trained networks to estimate whether an observation resembles the learned normal-data distribution:

Encode a real observation x as E(x), then reconstruct it as G(E(x)). section 16.2.3
Treat large reconstruction error as evidence that the encoder cannot represent the observation normally. section 16.2.3

Practical detector

PyOD's ALAD detector provides a concise implementation: instantiate ALAD, fit it to X, inspect decision_scores_, and optionally plot learning curves.

How Outlier Detection in Python covers it

liveBook manning.com

Read Outlier Detection in Python on liveBook Get the book at manning.com

Related concepts

adversarially learned anomaly detectionimplemented by · Adversarially learned anomaly detection is a GAN-based outlier-detection method using a bidirectional GAN architecture and reconstruction error scoring.
generative adversarial networkkind of · A generative adversarial network is a generative model with a generator and discriminator trained in competition, enabling the generator to produce samples resembling a target dataset or distribution.

This explainer was generated by Manning's knowledge graph from the text of Outlier Detection in Python. The section links point to the parts of the book it draws on.