Merge remote-tracking branch 'origin/master'

README.md

@@ -15,7 +15,7 @@ New algorithms such as Neural Fictitious Self-Play (NFSP)
 Regret Policy Gradients (RPG) [[4]](https://arxiv.org/pdf/1810.09026.pdf),
 Deep Counterfactual Regret Minimization (Deep CFR) [[5]](https://arxiv.org/pdf/1811.00164.pdf),
 and Single Deep CFR [[8]](https://arxiv.org/pdf/1901.07621.pdf)
-have recently combined Deep (Reinforcement) Learning conventional methods like CFR and Fictitious-Play to learn
+have recently combined Deep (Reinforcement) Learning with conventional methods like CFR and Fictitious-Play to learn
 approximate Nash equilibria while only ever visiting a fraction of the game's states.
 
 
@@ -235,8 +235,8 @@ information processing systems. 2009.
 [3] Heinrich, Johannes, and David Silver. "Deep reinforcement learning from self-play in imperfect-information games."
 arXiv preprint arXiv:1603.01121 (2016).
 
-[4] Jin, Peter H., Sergey Levine, and Kurt Keutzer. "Regret Minimization for Partially Observable Deep Reinforcement
-Learning." arXiv preprint arXiv:1710.11424 (2017).
+[4] Srinivasan, Sriram, et al. "Actor-critic policy optimization in partially observable multiagent environments."
+Advances in Neural Information Processing Systems. 2018.
 
 [5] Brown, Noam, et al. "Deep Counterfactual Regret Minimization." arXiv preprint arXiv:1811.00164 (2018).
 
@@ -256,4 +256,4 @@ information processing systems. 2008.
 [11] Tammelin, Oskari. "Solving large imperfect information games using CFR+." arXiv preprint arXiv:1407.5042 (2014).
 
 [12] Brown, Noam, and Tuomas Sandholm. "Solving Imperfect-Information Games via Discounted Regret Minimization."
-arXiv preprint arXiv:1809.04040 (2018).
+arXiv preprint arXiv:1809.04040 (2018).