Research Article
Research on Open Innovation Intelligent Decision-Making of Cross-Border E-Commerce Based on Federated Learning
Table 1
Federated learning averaging algorithm.
| Federated averaging algorithm |
| Input: | Randomly select M(n = 1,…,M)participants from N participants; I is the time of cycle; | is The initial gradient provided for the server; | E is the number of time periods on the local SGD (stochastic gradient descent); | Server: To initialize | From = 1,…,T: | The server sends to : | Each client : | ←Local Update(n, ) | ← | Local Update(n, ): | Number of local device sessions for each participant 1 to E: | Update according to the learning rate and | Feedback to the server |
| Output: | Optimized global gradient |
|
|