TinyTL: ReduceMemory, NotParameters forEfficientOn-DeviceLearning

Feb-9-2026, 04:14:59 GMT–Neural Information Processing Systems

Existing work solves this problem by reducing the number of trainable parameters. However,this doesn'tdirectly translate to memory saving since the major bottleneck is the activations, not parameters.

artificial intelligence, arxivpreprintarxiv, machine learning, (17 more...)

Neural Information Processing Systems

Feb-9-2026, 04:14:59 GMT

Conferences PDF

Add feedback

Country:
- North America
  - Canada > British Columbia
    - Metro Vancouver Regional District > Vancouver (0.04)
  - United States > Massachusetts
    - Middlesex County > Cambridge (0.04)

Technology:
- Information Technology > Artificial Intelligence > Machine Learning (1.00)

Duplicate Docs Excel Report

Title
TinyTL: Reduce Memory, Not Parameters for Efficient On-Device Learning

Similar Docs Excel Report more

Title	Similarity	Source
None found