Latent Variable and ELBO
From latent variables to the evidence lower bound.
Position in the series
latent variable -> ELBO
Outline
In this section, we will answer these questions:
- Why introduce latent variables?
- Marginal likelihood and the intractable integral.
- Variational posterior $q(z \mid x)$.
- Jensen’s inequality and ELBO.
- Reconstruction term and KL term.
- ELBO as a probabilistic modeling principle.
Notes
Since the distribution of real-world data is highly complex, we introduce latent variables $z$ to explain the observed data $x$. Thus, we do not model the data distribution directly as:
\[p_\theta(x)\]Instead, we assume the data is generated through latent variables $z$:
\[\begin{aligned} z &\sim p(z), \\ x &\sim p_\theta(x \mid z). \end{aligned}\]Finally,
\[p_\theta(x) = \int p_\theta(x \mid z)p(z)\,dz\]In this place, $z$ is the latent variable.