Initialization of ReLUs for Dynamical Isometry

Rebekka Burkholz, Alina Dubatovka

Aug-20-2025, 04:54:47 GMT–Neural Information Processing Systems

Deep learning relies critically on good parameter initialization prior to training.

artificial intelligence, machine learning, neural network, (18 more...)

Neural Information Processing Systems

Aug-20-2025, 04:54:47 GMT

Conferences PDF

Country:
- Europe
  - France > Hauts-de-France
    - Nord > Lille (0.04)
  - Italy > Sardinia (0.04)
  - Spain > Canary Islands (0.04)
  - Sweden > Stockholm
    - Stockholm (0.04)
  - Switzerland > Zürich
    - Zürich (0.04)
- North America
  - Canada > Alberta
    - Census Division No. 15 > Improvement District No. 9 > Banff (0.04)
  - Puerto Rico > San Juan
    - San Juan (0.04)
  - United States
    - California > Los Angeles County
      - Long Beach (0.04)
    - District of Columbia > Washington (0.04)
    - Georgia > Fulton County
      - Atlanta (0.04)
    - Massachusetts > Suffolk County
      - Boston (0.04)
- Oceania > Australia
  - New South Wales > Sydney (0.14)

Industry:
- Health & Medicine (0.68)

Technology:
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.68)

Duplicate Docs Excel Report

Title
Initialization of ReLUs for Dynamical Isometry

Similar Docs Excel Report more

Title	Similarity	Source
None found