@@ -14,9 +14,9 @@ As illustrated below the user specified probabilistic program specifies how to g
The goal is to infer a probabilistic model θ for observed data x<sub>1</sub>, ..., x<sub>n</sub>: p(θ|x<sub>1</sub>, ..., x<sub>n</sub>).
This is done by using Bayesian statistics and requires usually the following steps:
1. Define a probabilistic model (e.g. Bernulli distribution) by (i) setting up the parameter(s) for the model (θ; i.e. the latent random variable(s)) and (ii) seting up a (discrete) set of random variables sampled from the model (x<sub>1</sub>, ..., x<sub>n</sub>; i.e. observed variables) used to check the quality of the model θ.
2. Specify the measured data (i.e. import real world data)
3. Run Inference using the model and the data (e.g. by Monte Carlo Sampling)
1. Define a probabilistic model (e.g. Bernulli distribution) by setting up the parameter(s) for the model (θ; i.e. the latent random variable(s)) and sample a (discrete) set of random variables from the model (same dimension as the real world data and used to check the quality of the model θ). Assumptions are encoded with prior distributions over the variables of the model.
2. Specify the measured / observed data (i.e. import real world data)
3. Run Inference using the model and the data (e.g. by Monte Carlo Sampling) to acquire posterior distribution. In other words, the inference adjusts the prior distribution using the observed data data to give a more precise model.
**Conceptual Categorisation of Deep Probabilistic Programming Languages (PPL)**
