| Download | - View final version: Correspondence between neuroevolution and gradient descent (PDF, 1.2 MiB)
- View supplementary information: Correspondence between neuroevolution and gradient descent (PDF, 2.9 MiB)
|
|---|
| DOI | Resolve DOI: https://doi.org/10.1038/s41467-021-26568-2 |
|---|
| Author | Search for: Whitelam, StephenORCID identifier: https://orcid.org/0000-0002-0086-6803; Search for: Selin, Viktor; Search for: Park, Sang-Won; Search for: Tamblyn, Isaac1 |
|---|
| Affiliation | - National Research Council of Canada. Security and Disruptive Technologies
|
|---|
| Format | Text, Article |
|---|
| Subject | applied mathematics; statistical physics |
|---|
| Abstract | We show analytically that training a neural network by conditioned stochastic mutation or neuroevolution of its weights is equivalent, in the limit of small mutations, to gradient descent on the loss function in the presence of Gaussian white noise. Averaged over independent realizations of the learning process, neuroevolution is equivalent to gradient descent on the loss function. We use numerical simulation to show that this correspondence can be observed for finite mutations, for shallow and deep neural networks. Our results provide a connection between two families of neural-network training methods that are usually considered to be fundamentally different. |
|---|
| Publication date | 2021-11-02 |
|---|
| Publisher | Nature Research |
|---|
| Licence | |
|---|
| In | |
|---|
| Related data | |
|---|
| Language | English |
|---|
| Peer reviewed | Yes |
|---|
| Export citation | Export as RIS |
|---|
| Report a correction | Report a correction (opens in a new tab) |
|---|
| Record identifier | f11ecf04-e63f-4322-88bd-1a75618ec617 |
|---|
| Record created | 2022-02-18 |
|---|
| Record modified | 2022-02-22 |
|---|
