From 23d2945bf85d67b2a9bcd55ddccd3b54fbf736a0 Mon Sep 17 00:00:00 2001 From: Gal Leibovich Date: Sun, 19 Aug 2018 11:02:45 +0300 Subject: [PATCH] Update README.md --- benchmarks/dqn/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/benchmarks/dqn/README.md b/benchmarks/dqn/README.md index 8b28c83..d617aa3 100644 --- a/benchmarks/dqn/README.md +++ b/benchmarks/dqn/README.md @@ -1,7 +1,7 @@ # DQN Each experiment uses 3 seeds. -The parameters used for DQN are the same parameters as described in the [original paper](https://arxiv.org/abs/1607.05077.pdf). +The parameters used for DQN are the same parameters as described in the [original paper](https://www.cs.toronto.edu/~vmnih/docs/dqn.pdf), except for the optimizer (changed to ADAM) and learning rate (1e-4) used. ### Breakout DQN - single worker