Skip to content

Optimizer Selection for the ra_vf Network #8

@allen23456

Description

@allen23456

Hello, I would like to inquire why the code uses SGD as the optimizer for training the ra_vf network. After attempting to switch to Adam, the convergence speed increased.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions