ClassroomThe classroom
Back to the lesson libraryMECHANISM · 13 MIN
M31.5 CONNECT THE MECHANISM

Learn new information while measuring what is forgotten

A trail camera learns spring's birds and forgets autumn's deer. Watch a two-weight model forget, then fix it with replay and with a penalty that guards the weights that mattered.

LESSON OVERVIEW13 min lesson

Lesson overview

A trail camera learns spring's birds and forgets autumn's deer. Watch a two-weight model forget, then fix it with replay and with a penalty that guards the weights that mattered.

What you’ll explore

  • Measure forgetting across a task sequence, explain why shared weights forget, protect old skills with replay or an EWC penalty, and describe online learning by its regret.
GO TO THE SOURCE

Original explanations, connected to the research.

Overcoming catastrophic forgetting in neural networks (Kirkpatrick et al., 2017)Gradient Episodic Memory for Continual Learning (Lopez-Paz & Ranzato, 2017)Online Learning and Online Convex Optimization (Shalev-Shwartz, 2012)
Suggest a correction

A precise note can make an explanation better.

Choose the scene and describe what needs attention. Download a feedback file to share through a channel you already use. This page does not send feedback or connect you with a reviewer.

The file includes this note, the scene title, and lesson metadata. Your saved progress and quiz responses are excluded. Download before leaving or reloading to keep your note.