Clearly state and cite cutting edge optimization approaches to determine phase angles

demonstrations/tutorial_apply_qsvt.metadata.json

@@ -9,7 +9,7 @@
         }
     ],
     "dateOfPublication": "2023-08-22T00:00:00+00:00",
-    "dateOfLastModification": "2023-09-05T00:00:00+00:00",
+    "dateOfLastModification": "2023-09-28T00:00:00+00:00",
     "categories": [
         "Quantum Computing",
         "Algorithms",

demonstrations/tutorial_apply_qsvt.py

@@ -105,7 +105,7 @@ def my_circuit(phase_angles):
 #
 # Phase Angles from PyQSP
 # ^^^^^^^^^^^^^^^^^^^^^^^
-# There are many methods for computing the phase angles (see [#phaseeval]_,
+# There are many numerical methods for computing the phase angles (see
 # [#machineprecision]_, [#productdecomp]_). They can be readily used with
 # PennyLane as long as the convention used to define the rotations
 # matches the one used when applying QSVT. This is as simple as specifying the
@@ -271,9 +271,9 @@ def loss_func(phi):
 plt.show()
 
 ###############################################################################
-# Awesome, we successfully optimized the phase angles! While we used a standard loss function
-# and optimizer, users have the freedom to explore any optimizer, loss function, and sampling
-# scheme when training the phase angles for QSVT.
+# Awesome, we successfully optimized the phase angles! While we used a simple loss function
+# and optimizer, more sophisticated optimization schemes have been presented in literature to
+# robustly train the phase angles for QSVT (see [#phaseeval]_).
 #
 # Let :math:`\hat{U}_{qsvt}(\vec{\phi}, x)` represent the unitary matrix of the QSVT algorithm.
 # Both of the methods above produce phase angles :math:`\vec{\phi}` such that: