diff --git a/docs/getting-started/installation.rst b/docs/getting-started/installation.rst
index 437c8eef4..214709cd6 100644
--- a/docs/getting-started/installation.rst
+++ b/docs/getting-started/installation.rst
@@ -6,6 +6,21 @@ Prerequisites
 -------------
 
 - Python 3.8+
+- Specific versions of pip and setuptools due to \
+  `a bug with gym <https://github.com/openai/gym/issues/3176>`_:
+
+.. code-block:: bash
+
+    pip install -U setuptools==65.5.0 pip==21
+
+- (on ARM64 Macs) you need to set environment variables due to \
+  `a bug in grpcio <https://stackoverflow.com/questions/66640705/how-can-i-install-grpcio-on-an-apple-m1-silicon-laptop>`_:
+
+.. code-block:: bash
+
+    export GRPC_PYTHON_BUILD_SYSTEM_OPENSSL=1
+    export GRPC_PYTHON_BUILD_SYSTEM_ZLIB=1
+
 - (Optional) OpenGL (to render gym environments)
 - (Optional) FFmpeg (to encode videos of renders)
 - (Optional) MuJoCo (follow instructions to install `mujoco\_py v1.5 here`_)
diff --git a/docs/main-concepts/experts.rst b/docs/main-concepts/experts.rst
index aa9caeb92..cda38cd3f 100644
--- a/docs/main-concepts/experts.rst
+++ b/docs/main-concepts/experts.rst
@@ -12,7 +12,7 @@ learning library.
 For example, BC and DAgger can learn from an expert policy and the command line
 interface of AIRL/GAIL allows one to specify an expert to sample demonstrations from.
 
-In the :doc:`../getting-started/first-steps` tutorial, we first train an expert policy
+In the :doc:`../getting-started/first_steps` tutorial, we first train an expert policy
 using the stable-baselines3 library and then imitate it's behavior using
 :doc:`../algorithms/bc`.
 In practice, you may want to load a pre-trained policy for performance reasons.