From dfde990f0e6473dc17ed038f7bccc2b5ba7cae64 Mon Sep 17 00:00:00 2001
From: wenzhangliu <liu_wzh@foxmail.com>
Date: Tue, 16 Jan 2024 21:30:20 +0800
Subject: [PATCH] tutorials for new algorithms (#22)

---
 docs/source/documents/usage/tutorials_new.rst | 37 +++++++++++++++++++
 1 file changed, 37 insertions(+)
 create mode 100644 docs/source/documents/usage/tutorials_new.rst

diff --git a/docs/source/documents/usage/tutorials_new.rst b/docs/source/documents/usage/tutorials_new.rst
new file mode 100644
index 00000000..ba506ce2
--- /dev/null
+++ b/docs/source/documents/usage/tutorials_new.rst
@@ -0,0 +1,37 @@
+Tutorials for New Algorithms
+================================
+
+As a flexible deep reinforcement learning library, it is easy for newers to extend new algorithms to XuanCe
+and build customized algorithms based on XuanCe.
+As is mentioned in :doc:`here <../../index>`, an algorithm consists of four components,namely representation, policy, learner, and agent.
+Hence, one can create a customized algorithm by the following steps.
+
+.. raw:: html
+
+   <br><hr>
+
+Step 1: Customized Representation
+------------------------------------
+
+
+
+
+Step 2: Customized Policy
+------------------------------------
+
+
+
+
+
+Step 3: Customized Learner
+------------------------------------
+
+
+
+
+
+Step 4: Customized Agent
+------------------------------------
+
+
+