An Investigation of How Neural Networks Learn from the Experiences of Peers Through Periodic Weight Averaging

By Josh Bhagat Smith, Michael S. Gashler. 2017

16 citations recorded by the source corpus.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Sidak Pal Singh
2. Martin Jaggi
3. Mikhail Iu. Leontev
4. V. Islenteva
5. Sergey V. Sukhov
6. Simon Lacoste-Julien
7. Sebastian U. Stich
8. Peter Richtárik

Citations in the corpus

• Model Fusion via Optimal Transport (2019, background): “share a significant part of the training history [13, 14], by relying on SGD with periodic averaging, also called local SGD [16].”
• Non-iterative Knowledge Fusion in Deep Convolutional Neural Networks (2018, methodology): “Instead, the approach of [32, 33] would be ideal in this case and we can use weight averaging.”