Update README with dataset visualization colab.

PiperOrigin-RevId: 320332055

Author: Rishabh Agarwal

README.md

@@ -40,7 +40,6 @@ run the command:
 gsutil -m cp -R gs://atari-replay-datasets/dqn/[GAME_NAME]
 ```
 
-
 This data can be generated by running the online agents using
 [`batch_rl/baselines/train.py`](https://github.com/google-research/batch_rl/blob/master/batch_rl/baselines/train.py) for 200 million frames
 (standard protocol). Note that the dataset consists of approximately 50 million
@@ -58,6 +57,10 @@ agent's previous action again, instead of the agent's new action.
 [gcp_bucket]: https://console.cloud.google.com/storage/browser/atari-replay-datasets
 [project_page]: https://offline-rl.github.io
 
+### Dataset Visualization Colab
+A colab showing how to load the dataset using dopamine replay buffers and
+visualize the observations can be found [here](https://colab.research.google.com/drive/17JQzRpIrguMjmZN5bL-RYzDJD5GSV8MJ?usp=sharing).
+
 ## Asymptotic Performance of offline agents on Atari-replay dataset
 
 <div>
@@ -190,7 +193,7 @@ If you find this open source release useful, please reference in your paper:
 > An Optimistic Perspective on Offline Reinforcement Learning
 > *International Conference on Machine Learning (ICML)*.
 
-    @article{agarwal2019optimistic,
+    @article{agarwal2020optimistic,
       title={An Optimistic Perspective on Offline Reinforcement Learning},
       author={Agarwal, Rishabh and Schuurmans, Dale and Norouzi, Mohammad},
       journal={International Conference on Machine Learning},