diff --git a/docs/source/mushroom_rl.rl_utils.rst b/docs/source/mushroom_rl.rl_utils.rst
index 262bacfe..4a9be33f 100644
--- a/docs/source/mushroom_rl.rl_utils.rst
+++ b/docs/source/mushroom_rl.rl_utils.rst
@@ -1,3 +1,6 @@
+Reinforcement Learning utils
+============================
+
 Eligibility trace
 -----------------
 
diff --git a/mushroom_rl/features/basis/fourier.py b/mushroom_rl/features/basis/fourier.py
index 6a85bfb2..e96b7033 100644
--- a/mushroom_rl/features/basis/fourier.py
+++ b/mushroom_rl/features/basis/fourier.py
@@ -9,8 +9,8 @@ class FourierBasis:
     .. math::
         \sum \cos{\pi(X - m)/\Delta c}
 
-    where X is the input, m is the vector of the minumum input values (for each
-    dimensions) , \Delta is the vector of maximum
+    where :math:`X` is the input, m is the vector of the minumum input values (for each dimensions) , :math:`\Delta` is
+    the vector of differences between maximum and minumun values for the variables.
 
     """
     def __init__(self, low, delta, c, dimensions=None):
@@ -19,11 +19,10 @@ def __init__(self, low, delta, c, dimensions=None):
 
         Args:
             low (np.ndarray): vector of minimum values of the input variables;
-            delta (np.ndarray): vector of the maximum difference between two
-             values of the input variables, i.e. delta = high - low;
+            delta (np.ndarray): vector of the maximum difference between two values of the input variables, i.e.
+                delta = high - low;
             c (np.ndarray): vector of weights for the state variables;
-            dimensions (list, None): list of the dimensions of the input to be
-                considered by the feature.
+            dimensions (list, None): list of the dimensions of the input to be considered by the feature.
 
         """
         self._low = low
diff --git a/mushroom_rl/features/basis/gaussian_rbf.py b/mushroom_rl/features/basis/gaussian_rbf.py
index 1389a1e7..1eadfa19 100644
--- a/mushroom_rl/features/basis/gaussian_rbf.py
+++ b/mushroom_rl/features/basis/gaussian_rbf.py
@@ -9,7 +9,7 @@ class GaussianRBF:
     .. math::
         \sum \dfrac{(X_i - \mu_i)^2}{\sigma_i}
 
-    where X is the input, \mu is the mean vector and \sigma is the scale parameter vector.
+    where :math:`X` is the input, :math:`\mu` is the mean vector and :math:`\sigma` is the scale parameter vector.
 
     """
     def __init__(self, mean, scale, dimensions=None):
diff --git a/mushroom_rl/features/basis/polynomial.py b/mushroom_rl/features/basis/polynomial.py
index 3483dce4..7e37b970 100644
--- a/mushroom_rl/features/basis/polynomial.py
+++ b/mushroom_rl/features/basis/polynomial.py
@@ -9,7 +9,7 @@ class PolynomialBasis:
     .. math::
         \prod X_i^{d_i}
 
-    where X is the input and d is the vector of the exponents of the polynomial.
+    where :math:`X~ is the input and :math:`d` is the vector of the exponents of the polynomial.
 
     """
     def __init__(self, dimensions=None, degrees=None, low=None, high=None):
diff --git a/mushroom_rl/features/tensors/random_fourier_tensor.py b/mushroom_rl/features/tensors/random_fourier_tensor.py
index 6e911e98..a356f40c 100644
--- a/mushroom_rl/features/tensors/random_fourier_tensor.py
+++ b/mushroom_rl/features/tensors/random_fourier_tensor.py
@@ -15,8 +15,8 @@ class RandomFourierBasis(nn.Module):
         \sin{\dfrac{PX}{\nu}+\varphi}
 
 
-    where X is the input, m is the vector of the minimum input values (for each dimensions), :math: `\Delta` is the
-    vector of maximum variations of the input variables, i.e. delta = high - low;
+    where :math:`X` is the input, :math:`P` is a random weights matrix, :math:`\nu` is the bandwidth parameter and
+    :math:`\varphi` is a bias vector.
 
     These features have been presented in:
 
diff --git a/mushroom_rl/rl_utils/parameters.py b/mushroom_rl/rl_utils/parameters.py
index 1f9d7899..5dc28ee0 100644
--- a/mushroom_rl/rl_utils/parameters.py
+++ b/mushroom_rl/rl_utils/parameters.py
@@ -116,14 +116,14 @@ def initial_value(self):
 
 
 class LinearParameter(Parameter):
-    """
+    r"""
     This class implements a linearly changing parameter according to the number of times it has been used.
     The parameter changes following the formula:
 
     .. math::
-        v_n = \textrm{clip}(v_0 + \dfrac{v_{th} - v_0}}{n}, v_{th})
+        v_n = \textrm{clip}(v_0 + \dfrac{v_{th} - v_0}{n}, v_{th})
 
-    where :math: `v_0` is the initial value of the parameter,  :math: `n` is the number of steps and  :math: `v_{th}` is
+    where :math:`v_0` is the initial value of the parameter,  :math:`n` is the number of steps and  :math:`v_{th}` is
     the upper or lower threshold for the parameter.
 
     """
@@ -153,13 +153,13 @@ def _compute(self, *idx, **kwargs):
 
 
 class DecayParameter(Parameter):
-    """
+    r"""
     This class implements a decaying parameter. The decay follows the formula:
 
     .. math::
         v_n = \dfrac{v_0}{n^p}
 
-    where :math: `v_0` is the initial value of the parameter,  :math: `n` is the number of steps and  :math: `p` is an
+    where :math:`v_0` is the initial value of the parameter,  :math:`n` is the number of steps and  :math:`p` is an
     arbitrary exponent.
 
     """