global explanation
In plain wordsAn explanation of how an AI model behaves overall — which factors matter most across all its decisions, and in what way — rather than why it made one particular decision.
An explanation of a model's behavior across its whole input distribution: which features matter most on average, how predictions vary with each feature, what interactions exist, and what rules approximate the decision logic. Tools include global feature importancefeature importance, n.: A score for how much each input — age, income, location — matters to a model's predictions. It helps people understand, debug and simplify models, but can mislead if read as cause and effect. (permutation importance, mean absolute SHAPSHAP, abbr.: A popular method that explains a single prediction by fairly dividing it among the inputs, using a rule from game theory for splitting a team's winnings among its players. values), effect plots (partial dependencepartial dependence plot, n.: A chart showing how a model's average prediction changes as one input is varied while everything else stays as it is in the data — for example, predicted house price against floor area., accumulated local effectsaccumulated local effects, n.: A chart of how a model's predictions change with one input, built from small local comparisons so it stays reliable when inputs are related to each other, unlike simpler average-based charts.), global surrogate modelssurrogate model, n.: A simpler stand-in model trained to imitate a more complex one — to explain it, or to approximate a slow or costly calculation cheaply. such as a shallow decision tree trained to mimic the model, concept-level tests such as TCAVtesting with concept activation vectors, phr.: A method for asking whether a human idea, such as 'wheels', matters to an image model's decision, such as 'car', by finding the direction for that idea inside the model and measuring its effect., and, for neural networks, mechanistic interpretabilitymechanistic interpretability, n.: A branch of AI research that tries to reverse-engineer a neural network — working out, piece by piece, which internal parts represent which concepts and how they combine to produce behavior. analyses of internal circuits. For an interpretable modelinterpretable model, n.: A model simple or structured enough that a person can follow exactly how it turns inputs into a decision — like a short checklist or a points-based score — with no separate explanation tool needed., the model itself is the global explanation. Global summaries suit validation, scientific insight and regulatory documentation, but averaging can hide subgroups or individuals for whom the model behaves differently.
The global explanation said income dominated, yet for applicants under 25 the model leaned mostly on account age.
Origin
Rooted in 1990s rule-extraction work that sought overall descriptions of trained neural networks; the global/local contrast became standard in explainable-AI surveys from about 2018.
Usage note
A global explanation can be right on average and wrong for a specific case; pair it with local explanationslocal explanation, n.: An explanation of why an AI model gave one particular answer for one particular case — this patient, this loan — rather than how it behaves in general. when individual decisions matter.
Related
- local explanationlocal explanation, n.: An explanation of why an AI model gave one particular answer for one particular case — this patient, this loan — rather than how it behaves in general.
- partial dependence plotpartial dependence plot, n.: A chart showing how a model's average prediction changes as one input is varied while everything else stays as it is in the data — for example, predicted house price against floor area.
- surrogate modelsurrogate model, n.: A simpler stand-in model trained to imitate a more complex one — to explain it, or to approximate a slow or costly calculation cheaply.
- feature importancefeature importance, n.: A score for how much each input — age, income, location — matters to a model's predictions. It helps people understand, debug and simplify models, but can mislead if read as cause and effect.
- interpretable modelinterpretable model, n.: A model simple or structured enough that a person can follow exactly how it turns inputs into a decision — like a short checklist or a points-based score — with no separate explanation tool needed.
- mechanistic interpretabilitymechanistic interpretability, n.: A branch of AI research that tries to reverse-engineer a neural network — working out, piece by piece, which internal parts represent which concepts and how they combine to produce behavior.
- local explanationlocal explanation, n.: An explanation of why an AI model gave one particular answer for one particular case — this patient, this loan — rather than how it behaves in general.
Mentioned in
explainable AIexplainable AI, n.: The field and set of methods that let people understand why an AI system produced a particular output — for example, which factors led it to reject a loan application., local explanationlocal explanation, n.: An explanation of why an AI model gave one particular answer for one particular case — this patient, this loan — rather than how it behaves in general., partial dependence plotpartial dependence plot, n.: A chart showing how a model's average prediction changes as one input is varied while everything else stays as it is in the data — for example, predicted house price against floor area., surrogate modelsurrogate model, n.: A simpler stand-in model trained to imitate a more complex one — to explain it, or to approximate a slow or costly calculation cheaply., accumulated local effectsaccumulated local effects, n.: A chart of how a model's predictions change with one input, built from small local comparisons so it stays reliable when inputs are related to each other, unlike simpler average-based charts., individual conditional expectationindividual conditional expectation, n.: A chart that traces, for each example separately, what a model would predict as one input is swept across its range, exposing differences between examples that an overall average would hide., testing with concept activation vectorstesting with concept activation vectors, phr.: A method for asking whether a human idea, such as 'wheels', matters to an image model's decision, such as 'car', by finding the direction for that idea inside the model and measuring its effect.