What is degenerate feedback loop and how to detect and prevent it?

Question

The degenerate feedback loop often makes machine learning models fail, then how to detect and prevent it?

Answer 1

A degenerate feedback loop can happen when the predictions themselves influence the feedback, which, in turn, influences the next iteration of the model. More formally, a degenerate feedback loop is created when a system’s outputs are used to generate the system’s future inputs, which, in turn, influence the system’s future outputs. In ML, a system’s predictions can influence how users interact with the system, and because users’ interactions with the system are sometimes used as training data to the same system, degenerate feedback loops can occur and cause unintended consequences. Degenerate feedback loops are especially common in tasks with natural labels from users, such as recommender systems and ads click-through-rate prediction.

For recommender systems, usually these methods/metrics are utilized to detect degenerate feedback loop: aggregate diversity, average coverage of longtail items and hit rate against popularity.

Two methods are randomization and positional features. In the first way, you sacrifice user experience for diversity; while in the second, you encode the position information as numerical or bool features and let the model learn how the position influences the user interaction.

Inferences:

1. Chapter 8 of Designing Machine Learning Systems