liveBook · Manning

Pruning

As explained in AI Model Evaluation

A model optimization technique that removes parameters, connections, or structures judged unnecessary for the target task or inference.

Pruning removes parameters, connections, or structures judged unnecessary for a target task, reducing model size and potentially shortening inference time. It is useful when latency, responsiveness, or infrastructure efficiency matters in production. section 4.2.2 section 4.4.2

Evaluate pruning against the system constraints that shape user experience:

  • Measure model inference latency, because model size and architecture affect how quickly predictions are generated. section 4.2.2
  • Consider application load time separately: faster model inference can improve responsiveness, but the full application still includes frontend loading and dynamic content. section 4.4.2
  • Compare pruning with quantization, distillation, and hardware acceleration when choosing an optimization strategy. section 4.2.2 section 4.4.2

How AI Model Evaluation covers it

How deep it goes
Positions pruning as an optimization option but does not explain pruning algorithms, selection criteria, or implementation steps.
What you can do after
Apply pruning as one option for reducing model size and inference time in latency-sensitive applications.
How it connects
Connects pruning with quantization, distillation, model inference latency, model size, hardware accelerators, and application load time.

Also explained in

Compare how 11 books explain pruning →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI Model Evaluation. The section links point to the parts of the book it draws on.