pruning

Definition

Plain language

Deleting some of a neural network's weights to shrink it or change its behavior.

Also called: prune, magnitude pruning

Why it matters: It can shrink a model or strip out unwanted behavior with little cost to accuracy, making it both an efficiency and a security tool.

For example, deleting the smallest 10% of a network's weights can wipe out a hidden backdoor while leaving its normal performance almost untouched.

Heard on the show

“And the lifecycle: roster edits every ten tasks — fork a winner, merge near-duplicates, prune dead weight, spawn a fresh specialist for uncovered task types.”

Mentioned in 14 episodes

1. 200
   The One Mechanism That Turns Twenty AI Clones Into an Actual Team
2. 186
   How a Frozen Model Went From 2% to 77% on Physics Puzzles — Without Retraining
3. 181
   How to Backpropagate Blame Through a Team of Chatbots — And When It Backfires
4. 179
   How DeepSeek Made One User Faster Without Slowing Down the Crowd
5. 158
   How Floating-Point Rounding Lets a Model Tell Which Chip It's On — And Misbehave
6. 151
   Why More Experience Made This AI Agent Worse, And How to Fix It
7. 131
   Why Autonomous Research Agents Forget Their Own Lessons, and Arbor's Fix
8. 119
   Beating Reinforcement Learning Without Ever Touching the Model's Weights
9. 109
   An AI Got Caught Reading the Answer Key, And Why That Catch Matters
10. 106
    Giving Agents a Notebook Instead of New Weights: How ExpGraph Lets Frozen Models Learn
11. 067
    An AI Just Solved a 1996 Erdős Problem—and the Simplest Agent Won
12. 064
    When Agent Memory Stops Being a Database and Starts Being a Skill
13. 046
    When the AI Optimizer Edits the Grade Book: Why Harnessing Evolution Needs a Wall
14. 036
    Sparse Attention Was the Wrong Frame. Treat It as Geometry Instead.

Related concepts

Related terms