good robot paper better link name

Author: ahundt

README.md

@@ -96,7 +96,7 @@ Calibration
 Reinforcement Learning
 ----------------------
 
-- ["Good Robot!": Efficient Reinforcement Learning for Multi-Step Visual Tasks with Sim to Real Transfer](https://github.com/jhu-lcsr/good_robot) - A real robot completes multi-step tasks after <20k simulated actions. [paper](https://arxiv.org/abs/1909.11730) (disclaimer: @ahundt is first author)
+- ["Good Robot!": Efficient Reinforcement Learning for Multi-Step Visual Tasks with Sim to Real Transfer](https://github.com/jhu-lcsr/good_robot) - A real robot completes multi-step tasks after <20k simulated actions. [Good Robot on ArXiV](https://arxiv.org/abs/1909.11730) (disclaimer: @ahundt is first author)
 - [TensorForce](https://github.com/reinforceio/tensorforce) - A TensorFlow library for applied reinforcement learning
 - [gqcnn](https://github.com/BerkeleyAutomation/gqcnn) -  [Grasp Quality Convolutional Neural Networks (GQ-CNNs)](https://berkeleyautomation.github.io/gqcnn/info/info.html) for grasp planning using training datasets from the [Dexterity Network (Dex-Net)](https://berkeleyautomation.github.io/dex-net)
 - [Guided Policy Search](https://github.com/cbfinn/gps) - Guided policy search (gps) algorithm and LQG-based trajectory optimization, meant to help others understand, reuse, and build upon existing work.