Scoring Metrics

Problem Scopes

At Antigranular, we've focused our problem scopes on two of the most fundamental tasks in data science: regression and classification.

Regression: In this arena, we measure performance using L1 and L2 metrics. Regression seeks to predict a continuous outcome variable from one or more predictors. The L1 metric, also known as Mean Absolute Error (MAE), averages the absolute differences between predictions and actual values, whereas the L2 metric, Mean Squared Error (MSE), squares these differences before averaging. These metrics provide the flexibility to either penalise all errors equally (L1) or to give more weight to larger errors (L2).

Classification: In classification tasks, the goal is to predict a categorical outcome from a set of variables. Accuracy serves as our metric of choice for classification tasks. It's the simplest way of evaluating model performance, calculated as the number of correct predictions divided by the total number of predictions. This straightforward metric makes it easy for you to understand how well your model is performing and how much room there is for improvement.

Remember, at Antigranular, it's not just about getting it right. It's about finding the best solution whilst minimising your privacy budget expenditure. So, make your choices wisely, strategise, and happy modelling!