reSGLD proposes to simulate a high-temperature particle for exploration and a low-temperature particle for exploitation and allows them to swap simultaneously. Moreover, a correction term is included to avoid biases.
Source: Non-convex Learning via Replica Exchange Stochastic Gradient MCMCPaper | Code | Results | Date | Stars |
---|
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |