diff --git a/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py b/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py
index 3485e7afbf1..efe35ca096f 100644
--- a/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py
+++ b/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py
@@ -103,7 +103,7 @@ def train():
   with tf.name_scope('dropout'):
     keep_prob = tf.placeholder(tf.float32)
     tf.summary.scalar('dropout_keep_probability', keep_prob)
-    dropped = tf.nn.dropout(hidden1, keep_prob)
+    dropped = tf.nn.dropout(hidden1, rate=(1 - keep_prob))
 
   # Do not apply softmax activation yet, see below.
   y = nn_layer(dropped, 500, 10, 'layer2', act=tf.identity)
diff --git a/tensorflow/python/grappler/hierarchical_controller.py b/tensorflow/python/grappler/hierarchical_controller.py
index c8f2f4245bb..e39988d96b5 100644
--- a/tensorflow/python/grappler/hierarchical_controller.py
+++ b/tensorflow/python/grappler/hierarchical_controller.py
@@ -883,7 +883,7 @@ class HierarchicalController(Controller):
                                                  actions.read(i - 1))
       )
       if self.hparams.keep_prob is not None:
-        signal = nn_ops.dropout(signal, self.hparams.keep_prob)
+        signal = nn_ops.dropout(signal, rate=(1 - self.hparams.keep_prob))
       next_c, next_h = lstm(signal, prev_c, prev_h, w_lstm, forget_bias)
       query = math_ops.matmul(next_h, attn_w_2)
       query = array_ops.reshape(
diff --git a/tensorflow/python/ops/nn_test.py b/tensorflow/python/ops/nn_test.py
index df07721e5d3..9979ad9fff0 100644
--- a/tensorflow/python/ops/nn_test.py
+++ b/tensorflow/python/ops/nn_test.py
@@ -313,7 +313,7 @@ class DropoutTest(test_lib.TestCase):
     num_iter = 10
     for keep_prob in [0.1, 0.5, 0.8]:
       t = constant_op.constant(1.0, shape=[x_dim, y_dim], dtype=dtypes.float32)
-      dropout = nn_ops.dropout(t, keep_prob)
+      dropout = nn_ops.dropout(t, rate=(1 - keep_prob))
       final_count = 0
       self.assertEqual([x_dim, y_dim], dropout.get_shape())
       for _ in xrange(0, num_iter):
@@ -340,7 +340,7 @@ class DropoutTest(test_lib.TestCase):
     num_iter = 10
     for keep_prob in [0.1, 0.5, 0.8]:
       t = constant_op.constant(1.0, shape=[x_dim, y_dim], dtype=dtypes.float32)
-      dropout = nn_ops.dropout(t, keep_prob, noise_shape=[x_dim, 1])
+      dropout = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[x_dim, 1])
       self.assertEqual([x_dim, y_dim], dropout.get_shape())
       final_count = 0
       for _ in xrange(0, num_iter):
@@ -364,7 +364,7 @@ class DropoutTest(test_lib.TestCase):
     num_iter = 10
     for keep_prob in [0.1, 0.5, 0.8]:
       t = constant_op.constant(1.0, shape=[x_dim, y_dim], dtype=dtypes.float32)
-      dropout = nn_ops.dropout(t, keep_prob, noise_shape=[x_dim, 1])
+      dropout = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[x_dim, 1])
       self.assertEqual([x_dim, y_dim], dropout.get_shape())
       for _ in xrange(0, num_iter):
         value = self.evaluate(dropout)
@@ -409,7 +409,9 @@ class DropoutTest(test_lib.TestCase):
     keep_prob = 0.5
     x = constant_op.constant(1.0, shape=[x_dim, y_dim], dtype=dtypes.float32)
     dropout_x = nn_ops.dropout(
-        x, keep_prob, noise_shape=array_ops.placeholder(dtypes.int32))
+        x,
+        rate=(1 - keep_prob),
+        noise_shape=array_ops.placeholder(dtypes.int32))
     self.assertEqual(x.get_shape(), dropout_x.get_shape())
 
   def testPartialShapedDropout(self):
@@ -419,7 +421,7 @@ class DropoutTest(test_lib.TestCase):
     for keep_prob in [0.1, 0.5, 0.8]:
       t = constant_op.constant(1.0, shape=[x_dim, y_dim], dtype=dtypes.float32)
       # Set noise_shape=[None, 1] which means [x_dim, 1].
-      dropout = nn_ops.dropout(t, keep_prob, noise_shape=[None, 1])
+      dropout = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[None, 1])
       self.assertEqual([x_dim, y_dim], dropout.get_shape())
       final_count = 0
       for _ in xrange(0, num_iter):
@@ -478,22 +480,23 @@ class DropoutTest(test_lib.TestCase):
     keep_prob = 0.5
     t = constant_op.constant(1.0, shape=[x_dim, y_dim], dtype=dtypes.float32)
     with self.assertRaises(ValueError):
-      _ = nn_ops.dropout(t, keep_prob, noise_shape=[x_dim, y_dim + 10])
+      _ = nn_ops.dropout(
+          t, rate=(1 - keep_prob), noise_shape=[x_dim, y_dim + 10])
     with self.assertRaises(ValueError):
-      _ = nn_ops.dropout(t, keep_prob, noise_shape=[x_dim, y_dim, 5])
+      _ = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[x_dim, y_dim, 5])
     with self.assertRaises(ValueError):
-      _ = nn_ops.dropout(t, keep_prob, noise_shape=[x_dim + 3])
+      _ = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[x_dim + 3])
     with self.assertRaises(ValueError):
-      _ = nn_ops.dropout(t, keep_prob, noise_shape=[x_dim])
+      _ = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[x_dim])
     # test that broadcasting proceeds
-    _ = nn_ops.dropout(t, keep_prob, noise_shape=[y_dim])
-    _ = nn_ops.dropout(t, keep_prob, noise_shape=[1, y_dim])
-    _ = nn_ops.dropout(t, keep_prob, noise_shape=[x_dim, 1])
-    _ = nn_ops.dropout(t, keep_prob, noise_shape=[1, 1])
+    _ = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[y_dim])
+    _ = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[1, y_dim])
+    _ = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[x_dim, 1])
+    _ = nn_ops.dropout(t, rate=(1 - keep_prob), noise_shape=[1, 1])
 
   def testNoDropoutFast(self):
     x = array_ops.zeros((5,))
-    y = nn_ops.dropout(x, keep_prob=1)
+    y = nn_ops.dropout(x, rate=0)
     self.assertTrue(x is y)
 
     y = nn_ops.dropout_v2(x, rate=0)