> 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/problem-framing/intro.md). # Introduction The same table is a different problem under a different label regime, and there is no reinforcement-setting page here. The chapter first names the task, then walks the learning settings from weak and scarce labels to streams, few-shot data, and the choice of a model family. After this chapter the reader can name the task and the learning setting, and reinforcement learning is in [Reinforcement Learning](/decision-intelligence/reinforcement-learning.md). ## Tasks The first step is to see where a problem sits among machine learning setups. The [Overview](/problem-framing/types-of-machine-learning.md) maps how those setups differ by how much supervision and feedback they use, so you can place a problem before you pick a method. When one example carries more than one label, [Multi Label Classification](/problem-framing/multi-label-classification.md) covers the main ways to assign several labels per example, with problem-transformation recipes, metrics, and tooling. ## Learning settings Once the task is named, the label regime decides the setting. [Weakly Supervised](/problem-framing/weakly-supervised.md) is for labels that are incomplete, inaccurate, or inexact rather than fully supervised. [Label Algorithms](/problem-framing/label-algorithms.md) handles unbalanced labels, label propagation and spreading, and label noise, for when labels are scarce, noisy, or need to be spread across a graph of related examples. [Semi Supervised](/problem-framing/semi-supervised.md) is training with lots of unlabeled data and only a little labeled data. When labels are expensive, [Active Learning](/problem-framing/active-learning.md) chooses which examples a human labels next, with uncertainty and diversity sampling and query-by-committee. [Online Learning](/problem-framing/online-learning.md) is for models that update as labeled examples arrive in a stream, including non-stationary settings. [N-Shot Learning](/problem-framing/n-shot-learning.md) covers zero-, one-, and few-shot learning, adapting with only a handful of labeled examples per class. With the task and the setting known, [Model Families](/problem-framing/model-families.md) maps the problem shape to an algorithm family. --- # 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/problem-framing/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.