> For the complete documentation index, see [llms.txt](https://www.mlcompendium.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://www.mlcompendium.com/decision-intelligence/intro.md). # Introduction This chapter is how an action is chosen and how you know it worked for the business. It starts with experiments you plan and test, moves to decisions that keep being made as data arrives, and ends with causality when you can only observe. After it, the reader can design a test, a bandit, or a sequential decision, and run tracking lives in [Experiment management](/ai-engineering/experiment-management.md). ## Experimentation The first way to know an action worked is to plan a test so you learn from fewer runs. [Design Of Experiments](/decision-intelligence/design-of-experiments.md) is where that planning starts, with introductions and tutorials in Python, R, and Matlab. [DOE Tools](/decision-intelligence/doe-tools.md) is the tooling rather than the theory: generators and notebooks such as doePy and PyDoe. [Factorial Design](/decision-intelligence/factorial-design.md) is the factorial experiment as one DoE layout. Once the runs are planned, [Hypothesis Testing](/decision-intelligence/hypothesis-testing.md) is the statistical language behind the decision they feed. [A/B Testing](/decision-intelligence/a-b-testing.md) applies that language to online controlled experiments: whether a product change moved the metric that matters. [Multi Armed Bandits](/decision-intelligence/multi-armed-bandits.md) keeps allocating trials while learning which arm pays off, and [Contextual Bandits](/decision-intelligence/contextual-bandits.md) chooses actions using side information, not only arm history. ## Sequential decisions A bandit already decides again at every step, and the next pages make that the whole setting. [Active Learning Algorithms](/decision-intelligence/active-learning-algorithms.md) is active-style learning when labels arrive in large streams, starting with the Passive Aggressive classifier. [Incremental Learning](/decision-intelligence/incremental-learning.md) extends a model as new input arrives instead of training only once. [Follow the regularized leader](/decision-intelligence/follow-the-regularized-leader.md) is FTRL, an online learning algorithm. [Reinforcement Learning](/decision-intelligence/reinforcement-learning.md) is agents that learn by interacting for reward. ## Causality When you cannot run the test at all, the comparison has to come from observation. [Propensity Score Matching](/decision-intelligence/propensity-score-matching.md) is propensity score matching and propensity modeling for observational comparisons. [Root Cause Effects (RCE/RCA)](/decision-intelligence/root-cause-effects-rce-rca.md) is root-cause analysis: finding which faults or graph structure explain observed incidents. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://www.mlcompendium.com/decision-intelligence/intro.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.