Posts

Understanding Causal Learning

Image
Traditional causal model: It represents a traditional causal model with two main components: Observational Data (X) : These are the effects or outcomes that we can measure or observe directly. Explicit Causes (Y) : These are the known factors that we believe have a direct impact on the observational data. Latent Causes (E) : These are hidden or unknown factors that may also influence the observational data but are not directly observed. Structured causal learning mode: it is a complex and structured causal learning model, often used in machine learning, with several components: Explainer (P(E | X)) : This is a model that tries to uncover the latent causes (E) given the observational data (X). It's like looking for hidden factors that explain the observed outcomes. Reasoner (P(Y | X, E)) : Given the observational data and the explanations from the Explainer, the Reasoner tries to infer the explicit causes (Y). It's about understanding what the known factors might be given the da...
Image
  Causal Cooperative Nets: A New Framework for Causal Relationship Discovery By Jun Ho Park The algorithm has three main components: Explainer:   The explainer module takes an observation as input and produces a causal explanation of the observation.  This explanation can be in the form of a causal graph,  a set of causal rules,  or any other representation that captures the causal relationships between the variables in the observation. Reasoner:   The reasoner module takes an observation and a causal explanation as input and produces an inferred label for the observation.  This inferred label is based on the causal relationships in the explanation. Producer:  The producer module takes an inferred label and a causal explanation as input and generates a reconstructed observation.  This reconstructed observation is an attempt to recreate the original observation based on the inferred label and the causal relationships in the explanation. The al...