DLR-RM/stable-baselines3

[Enhancement]: Wrong gains for weight initialization

Open

#1,559 建立於 2023年6月16日

在 GitHub 查看
 (2 留言) (0 反應) (1 負責人)Python (6,550 star) (1,407 fork)batch import
enhancementhelp wanted

描述

Enhancement

The recommended gains for the weight init depend on the used activation function, see torch docs. However, as for now the used gains are statically implemented and always the same in ActorCriticPolicies. See here.

I recommend making the gains dependent on the activation function used(, i.e. probably mainly ReLU and tanh).

If you agree with this, I would like to implement it myself and PR.

Thanks and a good day!

To Reproduce

--

Relevant log output / Error message

--

System Info

--

Checklist

  • I have checked that there is no similar issue in the repo
  • I have read the documentation
  • I have provided a minimal working example to reproduce the bug
  • I've used the markdown code blocks for both code and stack traces.

貢獻者指南

技術棧
pythonpytorch
領域
machine learning
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
3
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
1-3 hours
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
needs maintainer response
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
Basic knowledge of PyTorchUnderstanding of weight initializationFamiliarity with stable baselines3 codebase
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
60
研究方向
The issue is in stable baselines3/common/policies.py around line 589 where gains are statically set. The goal is to dynamically set gains based on activation function, following PyTorch's recommendations. The contributor should look at how activation functions are selected in the policy network and adjust the initialization accordingly. No linked PRs or assignee comments are visible.
[Enhancement]: Wrong gains for weight initialization · DLR-RM/stable-baselines3#1559 | Good First Issue