What is a weight?
A value that a model multiplies by another value.
weight explained in plain English
A value that a model multiplies by another value. Training is the process of determining a model's ideal weights; inference is the process of using those learned weights to make predictions.
Imagine a linear model with two features. Suppose that training determines the following weights (and bias): - The bias, b, has a value of 2.2 - The weight, w1 associated with one feature is 1.5. - The weight, w2 associated with the other feature is 0.4. Now imagine an example with the following feature values: - The value of one feature, x1, is 6. - The value of the other feature, x2, is 10. This linear model uses the following formula to generate a prediction, y':
If a weight is 0, then the corresponding feature doesn't contribute to the model. For example, if w1 is 0, then the value of x1 is irrelevant. --- See Linear regression in Machine Learning Crash Course for more information.
Example
Practitioners refer to weight when building, training, or evaluating machine learning systems. It appears in research papers, product documentation, and technical discussions about AI capabilities and limitations.
People also read
- bias
1.
- bias (math) or bias term
An intercept or offset from an origin.
- confirmation bias
The tendency to search for, interpret, favor, and recall information in a way that confirms one's pre-existing beliefs or hypotheses.
- counterfactual fairness
A fairness metric that checks whether a classification model produces the same result for one individual as it does for another individual who is identical to the first, except with respect to one or more sensitive attributes.
- demographic parity
A fairness metric that is satisfied if the results of a model's classification are not dependent on a given sensitive attribute.
- discriminative model
A model that predicts labels from a set of one or more features.
- equality of opportunity
A fairness metric to assess whether a model is predicting the desirable outcome equally well for all values of a sensitive attribute.
- equalized odds
A fairness metric to assess whether a model is predicting outcomes equally well for all values of a sensitive attribute with respect to both the positive class and negative class—not just one class or the other exclusively.
- evaluation
The process of measuring a model's quality or comparing different models against each other.
- fairness metric
A mathematical definition of "fairness" that is measurable.