Baseline models refer to simple or rudimentary models utilized in machine learning. They serve as a point of comparison for more complex models developed during the machine learning tasks. They are out-of-the-box models that require minimal effort and are easy to implement but can set a minimum level of performance that more sophisticated models will need to exceed.
A baseline model helps users comprehend the lowest limit of acceptable model performance. In a typical machine learning process, researchers or developers often build the baseline model first, then compare its performance with more complex models.
For example, in a classification task, a baseline model might simply be a model that always predicts the most common class. In a regression task, a baseline model might be a model that always predicts the average value of the output variable.
The performance of these baseline models sets the minimum threshold that any subsequent models should surpass to have useful predictive value. If a complex model cannot outperform the baseline, it suggests that the complex model may not be well-suited for the task, or it might need further optimization.
Baseline models are also valuable when working with a new dataset or problem, as they provide a quick and rough estimate of the minimum expected performance, allowing developers to gauge the difficulty of a prediction problem.
Download this guide to delve into the most common LLM security risks and ways to mitigate them.
Lakera Guard protects your LLM applications from cybersecurity risks with a single line of code. Get started in minutes. Become stronger every day.
Several people are typing about AI/ML security. Come join us and 1000+ others in a chat that’s thoroughly SFW.