// Hacker Noon · 18 March 2026

Measuring Catastrophic Forgetting in AI

Catastrophic forgetting occurs when neural networks lose previously learned knowledge while acquiring new tasks. This section outlines three core ways to measure it: retention (performance drop on old tasks), relearning (speed of reacquisition), and activation overlap (shared internal representation...

Hacker Noon

@hacker-noon · Adam Optimizer

hackernoon.com

Hacker Noon@hacker-noon