diff --git a/es_distributed/policies.py b/es_distributed/policies.py
index b3cfcbd..3e46d38 100644
--- a/es_distributed/policies.py
+++ b/es_distributed/policies.py
@@ -181,10 +181,10 @@ def _make_net(self, o):
             aidx_na = bins(x, adim, len(acvals_k), 'out')  # values in [0, k-1]
             a = tf.gather_nd(
                 acvals_ak,
-                tf.concat(2, [
+                tf.concat([
                     tf.tile(np.arange(adim)[None, :, None], [tf.shape(aidx_na)[0], 1, 1]),
                     tf.expand_dims(aidx_na, -1)
-                ])  # (n,a,2)
+                ], 2)  # (n,a,2)
             )  # (n,a)
         elif ac_bin_mode == 'continuous':
             a = U.dense(x, adim, 'out', U.normc_initializer(0.01))
diff --git a/es_distributed/tf_util.py b/es_distributed/tf_util.py
index 4eebf6c..52a7191 100644
--- a/es_distributed/tf_util.py
+++ b/es_distributed/tf_util.py
@@ -28,7 +28,7 @@ def max(x, axis=None, keepdims=False):
 def min(x, axis=None, keepdims=False):
     return tf.reduce_min(x, reduction_indices=None if axis is None else [axis], keep_dims = keepdims)
 def concatenate(arrs, axis=0):
-    return tf.concat(axis, arrs)
+    return tf.concat(arrs, axis)
 def argmax(x, axis=None):
     return tf.argmax(x, dimension=axis)
 
@@ -179,8 +179,8 @@ def intprod(x):
 
 def flatgrad(loss, var_list):
     grads = tf.gradients(loss, var_list)
-    return tf.concat(0, [tf.reshape(grad, [numel(v)])
-        for (v, grad) in zip(var_list, grads)])
+    return tf.concat([tf.reshape(grad, [numel(v)])
+        for (v, grad) in zip(var_list, grads)], 0)
 
 class SetFromFlat(object):
     def __init__(self, var_list, dtype=tf.float32):
@@ -202,7 +202,7 @@ def __call__(self, theta):
 
 class GetFlat(object):
     def __init__(self, var_list):
-        self.op = tf.concat(0, [tf.reshape(v, [numel(v)]) for v in var_list])
+        self.op = tf.concat([tf.reshape(v, [numel(v)]) for v in var_list], 0)
     def __call__(self):
         return get_session().run(self.op)