Documentation nits (#756)

* Fix heading level in tutorial.

* Fix capitalization of algorithm article titles.

Author: ernestum

docs/algorithms/density.rst

@@ -1,7 +1,7 @@
 .. _density docs:
 
 =============================
-Density-based reward modeling
+Density-Based Reward Modeling
 =============================
 
 Density-based reward modeling is an inverse reinforcement learning (IRL) technique that assigns higher rewards

docs/algorithms/preference_comparisons.rst

@@ -1,7 +1,7 @@
 .. _preference comparisons docs:
 
 ======================
-Preference comparisons
+Preference Comparisons
 ======================
 
 The preference comparison algorithm learns a reward function from preferences between pairs of trajectories.

docs/tutorials/8_train_custom_env.ipynb

@@ -66,7 +66,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Step 2: create the environment"
+    "## Step 2: create the environment"
    ]
   },
   {
@@ -88,7 +88,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Step 2a (recommended): add the environment to the gym registry"
+    "### Step 2a (recommended): add the environment to the gym registry"
    ]
   },
   {
@@ -165,7 +165,7 @@
    "metadata": {},
    "source": [
     "\n",
-    "## Step 2b: directly use the environment\n",
+    "### Step 2b: directly use the environment\n",
     "\n",
     "Alternatively, we can directly initialize the environment by instantiating the class we created earlier, and handle all the additional logic ourselves."
    ]