From 34ae3194b4a15c57a8e5f2ae4d70191703a68f5a Mon Sep 17 00:00:00 2001
From: Peter Zhokhov <peterz@openai.com>
Date: Thu, 27 Sep 2018 12:51:43 -0700
Subject: [PATCH] add a note about DQN algorithms not performing well

---
 README.md | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/README.md b/README.md
index a9b7bf6..06b67f6 100644
--- a/README.md
+++ b/README.md
@@ -94,6 +94,10 @@ DQN with Atari is at this point a classics of benchmarks. To run the baselines i
 ```
 python -m baselines.run --alg=deepq --env=PongNoFrameskip-v4 --num_timesteps=1e6
 ```
+*NOTE:*
+The DQN-based algorithms currently do not get high scores on the Atari games 
+(see GitHub issue [431](https://github.com/openai/baselines/issues/431))
+We are currently investigating this and recommend users to instead use PPO2.
 
 ## Saving, loading and visualizing models
 The algorithms serialization API is not properly unified yet; however, there is a simple method to save / restore trained models.