What you will learn in Evaluating AI Systems
Assessing effectiveness, reliability, and safety
Chapter by chapter
Chapter 1: The importance of AI evaluation
The chapter introduction uses real-world AI failures to show why evaluating complex systems before deployment matters and why it is difficult.
What you learn in chapter 1 →Chapter 2: Framing and scoping the evaluation
The introduction explains why an AI evaluation needs a shared purpose and scope before a team chooses how to measure it.
What you learn in chapter 2 →Chapter 3: Picking evaluation metrics
The introduction explains how metrics turn evaluation goals into evidence and why choosing them requires more than picking a familiar measure.
What you learn in chapter 3 →Chapter 4: Acquiring evaluation data
This introduction explains what evaluation data can look like and why its suitability determines what an evaluation can reveal.
What you learn in chapter 4 →Chapter 5: Acquiring judgments
This introduction defines judgments, explains their role in evaluation metrics, and shows why judgment quality affects evaluation results.
What you learn in chapter 5 →Concepts explained in this book
- evaluation activityA step in the process of planning, executing, interpreting, or acting on an AI evaluation.
- metricA quantitative or scored measure used to assess a system against an evaluation criterion.
- evaluation datasetData on which an AI system is assessed and from which evaluation measures are computed.
- LLM-as-a-judgeAn AI system that assesses outputs or behavior according to evaluation criteria.
- evaluation taskAn evaluation task is an activity an AI system is expected to perform and whose outcomes can be assessed.
- evaluation criterionA stated condition or standard used to determine whether a system satisfies a decision requirement.
- synthetic dataArtificially generated data designed to represent selected characteristics or scenarios rather than collected directly from real-world events.
- evaluation goalThe intended purpose or decision that an evaluation is meant to support.
- artificial intelligence systemA system incorporating artificial intelligence whose real-world impact can depend on models, interfaces, data pipelines, human interaction, and policies.
- ground truthGround truth is the reference standard used to judge whether system outputs are correct, acceptable, or incorrect.
- automatic metricA metric computed directly from data by predefined formulas or algorithms without evaluative judgment during computation.
- evaluation objectThe AI system, subsystem, or component whose performance or properties are being assessed.
- execution-based metricA metric that assesses system performance through outputs being executed, used, or acted upon in a workflow, application, or environment.
- fidelityA measure of whether a response is faithful to its source information or supporting evidence.
- judgment-based metricA metric whose assessment depends on evaluative judgments involving interpretation, discretion, or normative reasoning.
- reference-based metricA metric that evaluates system outputs by comparing them with one or more reference answers.
- reference-free metricA metric that assesses system outputs without comparing them to a fixed ground truth.
- diversityThe degree of diversity among examples in a dataset.
- annotation artifactAn unintended pattern in evaluation data, arising from its creation or annotation, that can let a system score well without demonstrating the intended capability.
- binary judgmentA judgment that classifies an output as either meeting a criterion or not, with no intermediate states.
- metric stabilityThe degree to which a metric's value remains similar when the evaluation dataset changes slightly.
- sample sizeSample size is the number of observations or examples included in a study or evaluation.
- task definitionA task definition specifies the activity an AI system is expected to perform and the conditions that define its scope and success.
- judgment formatA form in which a judgment is expressed, such as a binary choice, a scalar value, or a comparison between outputs.
- absolute judgmentA method of assessment in which a judge evaluates an output independently, such as by assigning it a score on a scale.
- relative judgmentA method of assessment in which judges compare outputs with one another rather than assigning each an independent score.
- robustnessThe ability of a system to maintain suitable behavior under varied, unclear, or challenging inputs and conditions.
- benchmark datasetA dataset used as a common reference for evaluating system performance.
- spurious correlationA correlation between data features and labels that does not reflect the underlying relationship a system is meant to learn.
- interpretabilityThe extent to which a system’s behavior, outputs, or evaluation measures can be made understandable to people.
- scalar judgmentA judgment that assigns an output a value on a numerical or ordinal scale.
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.
