Limitations of the Empirical Fisher Approximation for Natural Gradient Descent

Oct-2-2025, 15:51:48 GMT–Neural Information Processing Systems

Several highly visible works have advocated an approximation known as the empirical Fisher, drawing connections between approximate second-order methods and heuristics like Adam.

artificial intelligence, fisher, machine learning, (13 more...)

Neural Information Processing Systems

Oct-2-2025, 15:51:48 GMT

Conferences PDF

Add feedback

Country:
- Asia
  - Japan > Kyūshū & Okinawa
    - Okinawa (0.04)
  - Middle East > Israel
    - Haifa District > Haifa (0.04)
- Europe
  - France > Hauts-de-France
    - Nord > Lille (0.04)
  - Germany > Baden-Württemberg
    - Tübingen Region > Tübingen (0.04)
  - Spain > Andalusia
    - Granada Province > Granada (0.04)
  - Sweden > Stockholm
    - Stockholm (0.05)
  - Switzerland > Vaud
    - Lausanne (0.04)
- North America
  - Canada
    - Alberta > Census Division No. 15
      - Improvement District No. 9 > Banff (0.04)
    - British Columbia (0.04)
    - Quebec > Montreal (0.04)
  - United States
    - California
      - Los Angeles County > Long Beach (0.04)
      - San Diego County > San Diego (0.04)
    - Georgia > Fulton County
      - Atlanta (0.04)
    - Indiana > Hamilton County
      - Fishers (0.04)
- Oceania > Australia
  - New South Wales > Sydney (0.04)

Technology:
- Information Technology > Artificial Intelligence
  - Machine Learning
    - Neural Networks (1.00)
    - Statistical Learning > Gradient Descent (0.53)
  - Representation & Reasoning (1.00)