# Continual Learning and Catastrophic Forgetting: EWC to MESU
Status: public
Confidence: medium (0.78) (verified)
Last verified: 2026-05-28
Generation: ai_structured

## TL;DR
Continual Learning and Catastrophic Forgetting: EWC to MESU: Continual learning studies how models can learn new tasks or data over time without catastrophically forgetting earlier knowledge.

## Core Explanation
Common strategies include regularizing important parameters, replaying or storing examples, and constraining updates so new learning does not erase previous tasks. Evaluation requires sequences of tasks rather than a single static train-test split.

## Further Reading

- [Overcoming catastrophic forgetting in neural networks](https://doi.org/10.1073/pnas.1611835114)
- [Gradient Episodic Memory for Continual Learning](https://arxiv.org/abs/1706.08840)
- [Learning without Forgetting](https://arxiv.org/abs/1606.09282)