Decision rule
A decision rule is a criterion or procedure that maps evidence, outcomes, or conditions to a choice or action.
Causal AI
A decision rule is a rule that selects an action based on a utility, loss, or outcome criterion.
- How it is used
- Chooses actions in causal decision systems, automated policies, public-health treatment regimes, and reinforcement-learning settings.
- How deep it goes
- Defines expected-utility, minimax, and softmax rules, then compares them through dominance and Bayesian admissibility.
- Examples
- Uses investment selection, automated production decisions, public-health regimes, inverse reinforcement learning, and Pyro-based Bayesian inference.
- What you can do after
- Evaluate/understand action-selection rules and assess admissibility when statistical estimation or latent variables affect decisions.
Ensemble Methods for Machine Learning
A human-readable if-then rule corresponding to a path from the root of a decision tree to a leaf.
- How it is used
- Interprets decision-tree classifications through readable conditions and population partitions.
- How deep it goes
- Explains root-to-leaf construction and coverage, while deferring detailed rule interpretation for deep trees.
- Examples
- Uses an Iris classifier with petal-width and petal-length thresholds expressed as Python-style if-then rules.
- What you can do after
- Evaluate decision-tree behavior by translating root-to-leaf paths into explicit classification conditions.
Read how Ensemble Methods for Machine Learning explains decision rule →
Related concepts
- policyalternative to · A policy is a rule, strategy, or mapping that governs or determines decisions, actions, permissions, or behavior under specified conditions.
- softmax decision ruleincludes · A softmax decision rule selects actions probabilistically according to a softmax transformation of their scores or utilities.
- decision treeproduced by · A decision tree is a branching structure that represents sequential tests or choices, with paths leading to outcomes, predictions, or actions.
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.