Catastrophic forgetting – Definition & Meaning

Definition of catastrophic forgetting

: the abrupt loss of previously learned abilities or knowledge when a neural network is trained on new data or tasks that overwrite parameters important to the earlier learning
- Fine-tuning the assistant only on legal documents caused catastrophic forgetting on its general question-answering benchmark.
- The continual-learning method replayed older examples to reduce catastrophic forgetting.

Origin & history

first known use1989

The problem was documented in late-1980s connectionist research, notably work by Michael McCloskey and Neal Cohen in 1989, and was also called catastrophic interference. It regained prominence as neural networks began being adapted sequentially to many tasks and domains.

Test yourself

Which of these is the meaning of catastrophic forgetting?

Related terms

continual learning an approach in which a machine-learning system learns from a sequence or stream of changing… fine-tuning further training of an existing model on a smaller, focused dataset so it gets better at a… self-distillation a form of knowledge distillation in which a model learns from outputs, representations… model collapse a gradual loss of quality and diversity that happens when models are trained on content… post-training everything done to a model after pre-training to make it useful and safe, such as instruction… neural network a computing system made of many simple connected units ("neurons") arranged in layers, which…

Cite this entry

"Catastrophic forgetting." AI Dictionary, Dadgogo, https://dadgogo.com/dictionary/catastrophic-forgetting/. Accessed 8 Oct. 2026.