1307 lines (1307 with data), 372.7 kB

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "_VCz6stSPtCC"
      },
      "outputs": [],
      "source": [
        "# This cell is added by sphinx-gallery\n",
        "# It can be customized to whatever you like\n",
        "%matplotlib inline\n",
        "# !pip install pennylane"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "exXC6t6_PtCD"
      },
      "source": [
        "Introduction to Geometric Quantum Machine Learning\n",
        "==================================================\n",
        "\n",
        "::: {.meta}\n",
        ":property=\\\"og:description\\\": Using the natural symmetries in a quantum\n",
        "learning problem can improve learning :property=\\\"og:image\\\":\n",
        "<https://pennylane.ai/qml/_images/equivariant_thumbnail.jpeg>\n",
        ":::\n",
        "\n",
        "::: {.related}\n",
        "tutorial\\_equivariant\\_graph\\_embedding A permutation equivariant graph\n",
        "embedding\n",
        ":::\n",
        "\n",
        "*Author: Richard East --- Posted: 18 October 2022.*\n",
        "\n",
        "Introduction\n",
        "------------\n",
        "\n",
        "Symmetries are at the heart of physics. Indeed in condensed matter and\n",
        "particle physics we often define a thing simply by the symmetries it\n",
        "adheres to. What does symmetry mean for those in machine learning? In\n",
        "this context the ambition is straightforward --- it is a means to reduce\n",
        "the parameter space and improve the trained model\\'s ability to\n",
        "sucessfully label unseen data, i.e., its ability to generalise.\n",
        "\n",
        "Suppose we have a learning task and the data we are learning from has an\n",
        "underlying symmetry. For example, consider a game of Noughts and Crosses\n",
        "(aka Tic-tac-toe): if we win a game, we would have won it if the board\n",
        "was rotated or flipped along any of the lines of symmetry. Now if we\n",
        "want to train an algorithm to spot the outcome of these games, we can\n",
        "either ignore the existence of this symmetry or we can somehow include\n",
        "it. The advantage of paying attention to the symmetry is it identifies\n",
        "multiple configurations of the board as \\'the same thing\\' as far as the\n",
        "symmetry is concerned. This means we can reduce our parameter space, and\n",
        "so the amount of data our algorithm must sift through is immediately\n",
        "reduced. Along the way, the fact that our learning model must encode a\n",
        "symmetry that actually exists in the system we are trying to represent\n",
        "naturally encourages our results to be more generalisable. The encoding\n",
        "of symmetries into our learning models is where the term *equivariance*\n",
        "will appear. We will see that demanding that certain symmetries are\n",
        "included in our models means that the mappings that make up our\n",
        "algorithms must be such that we could transform our input data with\n",
        "respect to a certain symmetry, then apply our mappings, and this would\n",
        "be the same as applying the mappings and then transforming the output\n",
        "data with the same symmetry. This is the technical property that gives\n",
        "us the name \\\"equavariant learning\\\".\n",
        "\n",
        "In classical machine learning, this area is often referred to as\n",
        "geometric deep learning (GDL) due to the traditional association of\n",
        "symmetry to the world of geometry, and the fact that these\n",
        "considerations usually focus on deep neural networks (see or for a broad\n",
        "introduction). We will refer to the quantum computing version of this as\n",
        "*quantum geometric machine learning* (QGML).\n",
        "\n",
        "Representation theory in circuits\n",
        "---------------------------------\n",
        "\n",
        "The first thing to discuss is how do we work with symmetries in the\n",
        "first place? The answer lies in the world of group representation\n",
        "theory.\n",
        "\n",
        "First, let\\'s define what we mean by a group:\n",
        "\n",
        "**Definition**: A group is a set $G$ together with a binary operation on\n",
        "$G$, here denoted $\\circ$, that combines any two elements $a$ and $b$ to\n",
        "form an element of $G$, denoted $a \\circ b$, such that the following\n",
        "three requirements, known as group axioms, are satisfied as follows:\n",
        "\n",
        "1.  **Associativity**: For all $a, b, c$ in $G$, one has\n",
        "    $(a \\circ b) \\circ c=a \\circ (b \\circ c)$.\n",
        "\n",
        "2.  \n",
        "\n",
        "    **Identity element**: There exists an element $e$ in $G$ such that, for every $a$ in $G$, one\n",
        "\n",
        "    :   has $e \\circ a=a$ and $a \\circ e=a$. Such an element is unique.\n",
        "        It is called the identity element of the group.\n",
        "\n",
        "3.  \n",
        "\n",
        "    **Inverse element**: For each $a$ in $G$, there exists an element $b$ in $G$\n",
        "\n",
        "    :   such that $a \\circ b=e$ and $b \\circ a=e$, where $e$ is the\n",
        "        identity element. For each $a$, the element $b$ is unique: it is\n",
        "        called the inverse of $a$ and is commonly denoted $a^{-1}$.\n",
        "\n",
        "With groups defined, we are in a position to articulate what a\n",
        "representation is: Let $\\varphi$ be a map sending $g$ in group $G$ to a\n",
        "linear map $\\varphi(g): V \\rightarrow V$, for some vector space $V$,\n",
        "which satisfies\n",
        "\n",
        "$$\\varphi\\left(g_{1} g_{2}\\right)=\\varphi\\left(g_{1}\\right) \\circ \\varphi\\left(g_{2}\\right) \\quad \\text { for all } g_{1}, g_{2} \\in G.$$\n",
        "\n",
        "The idea here is that just as elements in a group act on each other to\n",
        "reach further elements, i.e., $g\\circ h = k$, a representation sends us\n",
        "to a mapping acting on a vector space such that\n",
        "$\\varphi(g)\\circ \\varphi(h) = \\varphi(k)$. In this way we are\n",
        "representing the structure of the group as a linear map. For a\n",
        "representation, our mapping must send us to the general linear group\n",
        "$GL(n)$ (the space of invertible $n \\times n$ matrices with matrix\n",
        "multiplication as the group multiplication). Note how this is both a\n",
        "group, and by virtue of being a collection of invertible matrices, also\n",
        "a set of linear maps (they\\'re all invertble matrices that can act on\n",
        "row vectors). Fundamentally, representation theory is based on the\n",
        "prosaic observation that linear algebra is easy and group theory is\n",
        "abstract. So what if we can study groups via linear maps?\n",
        "\n",
        "Now due to the importance of unitarity in quantum mechnics, we are\n",
        "particularly interested in the unitary representations: representations\n",
        "where the linear maps are unitary matrices. If we can identify these\n",
        "then we will have a way to naturally encode groups in quantum circuits\n",
        "(which are mostly made up of unitary gates).\n",
        "\n",
        "![](../demonstrations/geometric_qml/sphere_equivariant.png){.align-center\n",
        "width=\"45.0%\"}\n",
        "\n",
        "How does all this relate to symmetries? Well, a large class of\n",
        "symmetries can be characterised as a group, where all the elements of\n",
        "the group leave some space we are considering unchanged. Let\\'s consider\n",
        "an example: the symmetries of a sphere. Now when we think of this\n",
        "symmetry we probably think something along the lines of \\\"it\\'s the same\n",
        "no matter how we rotate it, or flip it left to right, etc\\\". There is\n",
        "this idea of being invariant under some operation. We also have the idea\n",
        "of being able to undo these actions: if we rotate one way, we can rotate\n",
        "it back. If we flip the sphere right-to-left we can flip it\n",
        "left-to-right to get back to where we started (notice too all these\n",
        "inverses are unique). Trivially we can also do nothing. What exactly are\n",
        "we describing here? We have elements that correspond to an action on a\n",
        "sphere that can be inverted and for which there exists an identity. It\n",
        "is also trivially the case here that if we consider three operations a,\n",
        "b, c from the set of rotations and reflections of the sphere, that if we\n",
        "combine two of them together then\n",
        "$a\\circ (b \\circ c) = (a\\circ b) \\circ c$. The operations are\n",
        "associative. These features turn out to literally define a group!\n",
        "\n",
        "As we\\'ve seen the group in itself is a very abstract creature; this is\n",
        "why we look to its representations. The group labels what symmetries we\n",
        "care about, they tell us the mappings that our system is invariant\n",
        "under, and the unitary representations show us how those symmetries look\n",
        "on a particular space of unitary matrices. If we want to encode the\n",
        "structure of the symmeteries in a quantum circuit we must restrict our\n",
        "gates to being unitary representations of the group.\n",
        "\n",
        "There remains one question: *what is equivariance?* With our newfound\n",
        "knowledge of group representation theory we are ready to tackle this.\n",
        "Let $G$ be our group, and $V$ and $W$, with elements $v$ and $w$\n",
        "respectively, be vector spaces over some field $F$ with a map $f$\n",
        "between them. Suppose we have representations\n",
        "$\\varphi: G \\rightarrow GL(V)$ and $\\psi: G \\rightarrow GL(W)$.\n",
        "Furthermore, let\\'s write $\\varphi_g$ for the representation of $g$ as a\n",
        "linear map on $V$ and $\\psi_g$ as the same group element represented as\n",
        "a linear map on $W$ respectively. We call $f$ *equivariant* if\n",
        "\n",
        "$$f(\\varphi_g(v))=\\psi_g(f(v)) \\quad \\text { for all } g\\in G.$$\n",
        "\n",
        "The importance of such a map in machine learning is that if, for\n",
        "example, our neural network layers are equivariant maps then two inputs\n",
        "that are related by some intrinsic symmetry (maybe they are reflections)\n",
        "preserve this information in the outputs.\n",
        "\n",
        "Consider the following figure for example. What we see is a board with a\n",
        "cross in a certain square on the left and some numerical encoding of\n",
        "this on the right, where the 1 is where the X is in the number grid. We\n",
        "present an equivariant mapping between these two spaces with respect to\n",
        "a group action that is a rotation or a swap (here a $\\pi$ rotation). We\n",
        "can either apply a group action to the original grid and then map to the\n",
        "number grid, or we could map to the number grid and then apply the group\n",
        "action. Equivariance demands that the result of either of these\n",
        "procedures should be the same.\n",
        "\n",
        "![](../demonstrations/geometric_qml/equivariant-example.jpg){.align-center\n",
        "width=\"80.0%\"}\n",
        "\n",
        "Given the vast amount of input data required to train a neural network\n",
        "the principle that one can pre-encode known symmetry structures into the\n",
        "network allows us to learn better and faster. Indeed it is the reason\n",
        "for the success of convolutional neural networks (CNNs) for image\n",
        "analysis, where it is known they are equivariant with respect to\n",
        "translations. They naturally encode the idea that a picture of a dog is\n",
        "symmetrically related to the same picture slid to the left by n pixels,\n",
        "and they do this by having neural network layers that are equivariant\n",
        "maps. With our focus on unitary representations (and so quantum\n",
        "circuits) we are looking to extend this idea to quantum machine\n",
        "learning.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ugPilCpKPtCF"
      },
      "source": [
        "Noughts and Crosses\n",
        "===================\n",
        "\n",
        "Let\\'s look at the game of noughts and crosses, as inspired by. Two\n",
        "players take turns to place a O or an X, depending on which player they\n",
        "are, in a 3x3 grid. The aim is to get three of your symbols in a row,\n",
        "column, or diagonal. As this is not always possible depending on the\n",
        "choices of the players, there could be a draw. Our learning task is to\n",
        "take a set of completed games labelled with their outcomes and teach the\n",
        "algorithm to identify these correctly.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "daTwFY-GPtCG"
      },
      "source": [
        "This board of nine elements has the symmetry of the square, also known\n",
        "as the *dihedral group*. This means it is symmetric under\n",
        "$\\frac{\\pi}{2}$ rotations and flips about the lines of symmetry of a\n",
        "square (vertical, horizontal, and both diagonals).\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pXOW9XfePtCG"
      },
      "source": [
        "![](../demonstrations/geometric_qml/NandC_sym.png){.align-center\n",
        "width=\"70.0%\"}\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "81LgOuSiPtCG"
      },
      "source": [
        "**The question is, how do we encode this in our QML problem?**\n",
        "\n",
        "First, let us encode this problem classically. We will consider a\n",
        "nine-element vector $v$, each element of which identifies a square of\n",
        "the board. The entries themselves can be $+1$,$0$,$-1,$ representing a\n",
        "nought, no symbol, or a cross. The label is one-hot encoded in a vector\n",
        "$y=(y_O,y_- , y_X)$ with $+1$ in the correct label and $-1$ in the\n",
        "others. For instance (-1,-1,1) would represent an X in the relevant\n",
        "position.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EbaXa7F7PtCG"
      },
      "source": [
        "To create the quantum model let us take nine qubits and let them\n",
        "represent squares of our board. We\\'ll initialise them all as\n",
        "$|0\\rangle$, which we note leaves the board invariant under the\n",
        "symmetries of the problem (flip and rotate all you want, it\\'s still\n",
        "going to be zeroes whatever your mapping). We will then look to apply\n",
        "single qubit $R_x(\\theta)$ rotations on individual qubits, encoding each\n",
        "of the possibilities in the board squares at an angle of\n",
        "$\\frac{2\\pi}{3}$ from each other. For our parameterised gates we will\n",
        "have a single-qubit $R_x(\\theta_1)$ and $R_y(\\theta_2)$ rotation at each\n",
        "point. We will then use $CR_y(\\theta_3)$ for two-qubit entangling gates.\n",
        "This implies that, for each encoding, crudely, we\\'ll need 18\n",
        "single-qubit rotation parameters and $\\binom{9}{2}=36$ two-qubit gate\n",
        "rotations. Let\\'s see how, by using symmetries, we can reduce this.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CS1oXl5EPtCG"
      },
      "source": [
        "![..](../demonstrations/geometric_qml/grid.jpg){.align-center\n",
        "width=\"35.0%\"}\n",
        "\n",
        "The indexing of our game board.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3lyfUhwJPtCH"
      },
      "source": [
        "The secret will be to encode the symmetries into the gate set so the\n",
        "observables we are interested in inherently respect the symmetries. How\n",
        "do we do this? We need to select the collections of gates that commute\n",
        "with the symmetries. In general, we can use the twirling formula for\n",
        "this:\n",
        "\n",
        "::: {.tip}\n",
        "::: {.title}\n",
        "Tip\n",
        ":::\n",
        "\n",
        "Let $\\mathcal{S}$ be the group that encodes our symmetries and $U$ be a\n",
        "unitary representation of $\\mathcal{S}$. Then,\n",
        "\n",
        "$$\\mathcal{T}_{U}[X]=\\frac{1}{|\\mathcal{S}|} \\sum_{s \\in \\mathcal{S}} U(s) X U(s)^{\\dagger}$$\n",
        "\n",
        "defines a projector onto the set of operators commuting with all\n",
        "elements of the representation, i.e.,\n",
        "$\\left[\\mathcal{T}_{U}[X], U(s)\\right]=$ 0 for all $X$ and\n",
        "$s \\in \\mathcal{S}$.\n",
        ":::\n",
        "\n",
        "The twirling process applied to an arbitrary unitary will give us a new\n",
        "unitary that commutes with the group as we require. We remember that\n",
        "unitary gates typically have the form $W = \\exp(-i\\theta H)$, where $H$\n",
        "is a Hermitian matrix called a *generator*, and $\\theta$ may be fixed or\n",
        "left as a free parameter. A recipe for creating a unitary that commutes\n",
        "with our symmetries is to *twirl the generator of the gate*, i.e., we\n",
        "move from the gate $W = \\exp(-i\\theta H)$ to the gate\n",
        "$W' = \\exp(-i\\theta\\mathcal{T}_U[H])$. When each term in the twirling\n",
        "formula acts on different qubits, then this unitary would further\n",
        "simplify to\n",
        "\n",
        "$$W' = \\bigotimes_{s\\in\\mathcal{S}}U(s)\\exp(-i\\tfrac{\\theta}{\\vert\\mathcal{S}\\vert})U(s)^\\dagger.$$\n",
        "\n",
        "For simplicity, we can absorb the normalization factor\n",
        "$\\vert\\mathcal{S}\\vert$ into the free parameter $\\theta$.\n",
        "\n",
        "So let\\'s look again at our choice of gates: single-qubit $R_x(\\theta)$\n",
        "and $R_y(\\theta)$ rotations, and entangling two-qubit $CR_y(\\phi)$\n",
        "gates. What will we get by twirling these?\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bbVT5h2HPtCH"
      },
      "source": [
        "In this particular instance we can see the action of the twirling\n",
        "operation geometrically as the symmetries involved are all permutations.\n",
        "Let\\'s consider the $R_x$ rotation acting on one qubit. Now if this\n",
        "qubit is in the centre location on the grid, then we can flip around any\n",
        "symmetry axis we like, and this operation leaves the qubit invariant, so\n",
        "we\\'ve identified one equivariant gate immediately. If the qubit is on\n",
        "the corners, then the flipping will send this qubit rotation to each of\n",
        "the other corners. Similarly, if a qubit is on the central edge then the\n",
        "rotation gate will be sent round the other edges. So we can see that the\n",
        "twirling operation is a sum over all the possible outcomes of performing\n",
        "the symmetry action (the sum over the symmetry group actions). Having\n",
        "done this we can see that for a single-qubit rotation the invariant maps\n",
        "are rotations on the central qubit, at all the corners, and at all the\n",
        "central edges (when their rotation angles are fixed to be the same).\n",
        "\n",
        "As an example consider the following figure, where we take a $R_x$ gate\n",
        "in the corner and then apply all the symmetries of a square. The result\n",
        "of this twirling leads us to have the same gate at all the corners.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Qj-0MV6ePtCH"
      },
      "source": [
        "![](../demonstrations/geometric_qml/twirl.jpeg){.align-center\n",
        "width=\"70.0%\"}\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YkTducrePtCH"
      },
      "source": [
        "For entangling gates the situation is similar. There are three invariant\n",
        "classes, the centre entangled with all corners, with all edges, and the\n",
        "edges paired in a ring.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wzz4R-IxPtCH"
      },
      "source": [
        "The prediction of a label is obtained via a one-hot-encoding by\n",
        "measuring the expectation values of three invariant observables:\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UuFwqycTPtCH"
      },
      "source": [
        "$$O_{-}=Z_{\\text {middle }}=Z_{4}$$\n",
        "\n",
        "$$O_{\\circ}=\\frac{1}{4} \\sum_{i \\in \\text { corners }} Z_{i}=\\frac{1}{4}\\left[Z_{0}+Z_{2}+Z_{6}+Z_{8}\\right]$$\n",
        "\n",
        "$$O_{\\times}=\\frac{1}{4} \\sum_{i \\in \\text { edges }} Z_{i}=\\frac{1}{4}\\left[Z_{1}+Z_{3}+Z_{5}+Z_{7}\\right]$$\n",
        "\n",
        "$$\\hat{\\boldsymbol{y}}=\\left(\\left\\langle O_{\\circ}\\right\\rangle,\\left\\langle O_{-}\\right\\rangle,\\left\\langle O_{\\times}\\right\\rangle\\right)$$\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5STgZUP6PtCI"
      },
      "source": [
        "This is the quantum encoding of the symmetries into a learning problem.\n",
        "A prediction for a given data point will be obtained by selecting the\n",
        "class for which the observed expectation value is the largest.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WSH0J4_XPtCI"
      },
      "source": [
        "Now that we have a specific encoding and have decided on our observables\n",
        "we need to choose a suitable cost function to optimise. We will use an\n",
        "$l_2$ loss function acting on pairs of games and labels $D={(g,y)}$,\n",
        "where $D$ is our dataset.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BNyvZ0HRPtCI"
      },
      "source": [
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ePThXcVlPtCI"
      },
      "source": [
        "Let\\'s now implement this!\n",
        "\n",
        "First let\\'s generate some games. Here we are creating a small program\n",
        "that will play Noughts and Crosses against itself in a random fashion.\n",
        "On completion, it spits out the winner and the winning board, with\n",
        "noughts as +1, draw as 0, and crosses as -1. There are 26,830 different\n",
        "possible games but we will only sample a few hundred.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "fBqEnXDYPtCI"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import random\n",
        "\n",
        "# Fix seeds for reproducability\n",
        "torch.backends.cudnn.deterministic = True\n",
        "torch.manual_seed(16)\n",
        "random.seed(16)\n",
        "\n",
        "#  create an empty board\n",
        "def create_board():\n",
        "    return torch.tensor([[0, 0, 0], [0, 0, 0], [0, 0, 0]])\n",
        "\n",
        "\n",
        "# Check for empty places on board\n",
        "def possibilities(board):\n",
        "    l = []\n",
        "    for i in range(len(board)):\n",
        "        for j in range(3):\n",
        "            if board[i, j] == 0:\n",
        "                l.append((i, j))\n",
        "    return l\n",
        "\n",
        "\n",
        "# Select a random place for the player\n",
        "def random_place(board, player):\n",
        "    selection = possibilities(board)\n",
        "    current_loc = random.choice(selection)\n",
        "    board[current_loc] = player\n",
        "    return board\n",
        "\n",
        "\n",
        "# Check if there is a winner by having 3 in a row\n",
        "def row_win(board, player):\n",
        "    for x in range(3):\n",
        "        lista = []\n",
        "        win = True\n",
        "\n",
        "        for y in range(3):\n",
        "            lista.append(board[x, y])\n",
        "\n",
        "            if board[x, y] != player:\n",
        "                win = False\n",
        "\n",
        "        if win:\n",
        "            break\n",
        "\n",
        "    return win\n",
        "\n",
        "\n",
        "# Check if there is a winner by having 3 in a column\n",
        "def col_win(board, player):\n",
        "    for x in range(3):\n",
        "        win = True\n",
        "\n",
        "        for y in range(3):\n",
        "            if board[y, x] != player:\n",
        "                win = False\n",
        "\n",
        "        if win:\n",
        "            break\n",
        "\n",
        "    return win\n",
        "\n",
        "\n",
        "# Check if there is a winner by having 3 along a diagonal\n",
        "def diag_win(board, player):\n",
        "    win1 = True\n",
        "    win2 = True\n",
        "    for x, y in [(0, 0), (1, 1), (2, 2)]:\n",
        "        if board[x, y] != player:\n",
        "            win1 = False\n",
        "\n",
        "    for x, y in [(0, 2), (1, 1), (2, 0)]:\n",
        "        if board[x, y] != player:\n",
        "            win2 = False\n",
        "\n",
        "    return win1 or win2\n",
        "\n",
        "\n",
        "# Check if the win conditions have been met or if a draw has occurred\n",
        "def evaluate_game(board):\n",
        "    winner = None\n",
        "    for player in [1, -1]:\n",
        "        if row_win(board, player) or col_win(board, player) or diag_win(board, player):\n",
        "            winner = player\n",
        "\n",
        "    if torch.all(board != 0) and winner == None:\n",
        "        winner = 0\n",
        "\n",
        "    return winner\n",
        "\n",
        "\n",
        "# Main function to start the game\n",
        "def play_game():\n",
        "    board, winner, counter = create_board(), None, 1\n",
        "    while winner == None:\n",
        "        for player in [1, -1]:\n",
        "            board = random_place(board, player)\n",
        "            counter += 1\n",
        "            winner = evaluate_game(board)\n",
        "            if winner != None:\n",
        "                break\n",
        "\n",
        "    return [board.flatten(), winner]\n",
        "\n",
        "\n",
        "def create_dataset(size_for_each_winner):\n",
        "    game_d = {-1: [], 0: [], 1: []}\n",
        "\n",
        "    while min([len(v) for k, v in game_d.items()]) < size_for_each_winner:\n",
        "        board, winner = play_game()\n",
        "        if len(game_d[winner]) < size_for_each_winner:\n",
        "            game_d[winner].append(board)\n",
        "\n",
        "    res = []\n",
        "    for winner, boards in game_d.items():\n",
        "        res += [(board, winner) for board in boards]\n",
        "\n",
        "    return res\n",
        "\n",
        "\n",
        "NUM_TRAINING = 450\n",
        "NUM_VALIDATION = 600\n",
        "\n",
        "# Create datasets but with even numbers of each outcome\n",
        "with torch.no_grad():\n",
        "    dataset = create_dataset(NUM_TRAINING // 3)\n",
        "    dataset_val = create_dataset(NUM_VALIDATION // 3)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xB-FbtShPtCJ"
      },
      "source": [
        "Now let\\'s create the relevant circuit expectation values that respect\n",
        "the symmetry classes we defined over the single-site and two-site\n",
        "measurements.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "B3_k2h_IPtCJ",
        "outputId": "49afcd58-5e30-48fa-b6ec-de2d62a41d03"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 3700x1000 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "import pennylane as qml\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Set up a nine-qubit system\n",
        "dev = qml.device(\"default.qubit.torch\", wires=9)\n",
        "\n",
        "ob_center = qml.PauliZ(4)\n",
        "ob_corner = (qml.PauliZ(0) + qml.PauliZ(2) + qml.PauliZ(6) + qml.PauliZ(8)) * (1 / 4)\n",
        "ob_edge = (qml.PauliZ(1) + qml.PauliZ(3) + qml.PauliZ(5) + qml.PauliZ(7)) * (1 / 4)\n",
        "\n",
        "# Now let's encode the data in the following qubit models, first with symmetry\n",
        "@qml.qnode(dev, interface=\"torch\")\n",
        "def circuit(x, p):\n",
        "    qml.Hadamard(wires=0)\n",
        "    qml.Hadamard(wires=1)\n",
        "    qml.Hadamard(wires=2)\n",
        "    qml.Hadamard(wires=3)\n",
        "    qml.Hadamard(wires=4)\n",
        "    qml.Hadamard(wires=5)\n",
        "    qml.Hadamard(wires=6)\n",
        "    qml.Hadamard(wires=7)\n",
        "    qml.Hadamard(wires=8)\n",
        "    qml.RY(x[0], wires=0)\n",
        "    qml.RY(x[1], wires=1)\n",
        "    qml.RY(x[2], wires=2)\n",
        "    qml.RY(x[3], wires=3)\n",
        "    qml.RY(x[4], wires=4)\n",
        "    qml.RY(x[5], wires=5)\n",
        "    qml.RY(x[6], wires=6)\n",
        "    qml.RY(x[7], wires=7)\n",
        "    qml.RY(x[8], wires=8)\n",
        "\n",
        "    # Centre single-qubit rotation\n",
        "    qml.RX(p[0], wires=4)\n",
        "    qml.RY(p[1], wires=4)\n",
        "\n",
        "    # Corner single-qubit rotation\n",
        "    qml.RX(p[2], wires=0)\n",
        "    qml.RX(p[2], wires=2)\n",
        "    qml.RX(p[2], wires=6)\n",
        "    qml.RX(p[2], wires=8)\n",
        "\n",
        "    qml.RY(p[3], wires=0)\n",
        "    qml.RY(p[3], wires=2)\n",
        "    qml.RY(p[3], wires=6)\n",
        "    qml.RY(p[3], wires=8)\n",
        "\n",
        "    # Edge single-qubit rotation\n",
        "    qml.RX(p[4], wires=1)\n",
        "    qml.RX(p[4], wires=3)\n",
        "    qml.RX(p[4], wires=5)\n",
        "    qml.RX(p[4], wires=7)\n",
        "\n",
        "    qml.RY(p[5], wires=1)\n",
        "    qml.RY(p[5], wires=3)\n",
        "    qml.RY(p[5], wires=5)\n",
        "    qml.RY(p[5], wires=7)\n",
        "\n",
        "    # Double Entagling two-qubit gates\n",
        "    # circling the edge of the board\n",
        "    qml.CRY(p[6], wires=[0, 1])\n",
        "    qml.CRY(p[6], wires=[2, 1])\n",
        "    qml.CRY(p[6], wires=[2, 5])\n",
        "    qml.CRY(p[6], wires=[8, 5])\n",
        "    qml.CRY(p[6], wires=[8, 7])\n",
        "    qml.CRY(p[6], wires=[6, 7])\n",
        "    qml.CRY(p[6], wires=[6, 3])\n",
        "    qml.CRY(p[6], wires=[0, 3])\n",
        "\n",
        "    # To the corners from the centre\n",
        "    qml.CRY(p[7], wires=[4, 0])\n",
        "    qml.CRY(p[7], wires=[4, 2])\n",
        "    qml.CRY(p[7], wires=[4, 6])\n",
        "    qml.CRY(p[7], wires=[4, 8])\n",
        "\n",
        "    # To the centre from the edges\n",
        "    qml.CRY(p[8], wires=[1, 4])\n",
        "    qml.CRY(p[8], wires=[3, 4])\n",
        "    qml.CRY(p[8], wires=[5, 4])\n",
        "    qml.CRY(p[8], wires=[7, 4])\n",
        "\n",
        "    # circling the edge of the board\n",
        "    qml.CRY(p[6], wires=[0, 1])\n",
        "    qml.CRY(p[6], wires=[2, 1])\n",
        "    qml.CRY(p[6], wires=[2, 5])\n",
        "    qml.CRY(p[6], wires=[8, 5])\n",
        "    qml.CRY(p[6], wires=[8, 7])\n",
        "    qml.CRY(p[6], wires=[6, 7])\n",
        "    qml.CRY(p[6], wires=[6, 3])\n",
        "    qml.CRY(p[6], wires=[0, 3])\n",
        "\n",
        "    # To the corners from the centre\n",
        "    qml.CRY(p[7], wires=[4, 0])\n",
        "    qml.CRY(p[7], wires=[4, 2])\n",
        "    qml.CRY(p[7], wires=[4, 6])\n",
        "    qml.CRY(p[7], wires=[4, 8])\n",
        "\n",
        "    # To the centre from the edges\n",
        "    qml.CRY(p[8], wires=[1, 4])\n",
        "    qml.CRY(p[8], wires=[3, 4])\n",
        "    qml.CRY(p[8], wires=[5, 4])\n",
        "    qml.CRY(p[8], wires=[7, 4])\n",
        "\n",
        "    return [qml.expval(ob_center), qml.expval(ob_corner), qml.expval(ob_edge)]\n",
        "\n",
        "\n",
        "fig, ax = qml.draw_mpl(circuit)([0] * 9, 18 * [0])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EZKg8hC1PtCJ"
      },
      "source": [
        "Let\\'s also look at the same series of gates but this time they are\n",
        "applied independently from one another, so we won\\'t be preserving the\n",
        "symmetries with our gate operations. Practically this also means more\n",
        "parameters, as previously groups of gates were updated together.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "Bhp9RWhXPtCJ",
        "outputId": "24413c39-e6ee-414b-e7c0-517743512713"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 3700x1000 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "@qml.qnode(dev, interface=\"torch\")\n",
        "def circuit_no_sym(x, p):\n",
        "\n",
        "    qml.Hadamard(wires=0)\n",
        "    qml.Hadamard(wires=1)\n",
        "    qml.Hadamard(wires=2)\n",
        "    qml.Hadamard(wires=3)\n",
        "    qml.Hadamard(wires=4)\n",
        "    qml.Hadamard(wires=5)\n",
        "    qml.Hadamard(wires=6)\n",
        "    qml.Hadamard(wires=7)\n",
        "    qml.Hadamard(wires=8)\n",
        "    qml.RY(x[0], wires=0)\n",
        "    qml.RY(x[1], wires=1)\n",
        "    qml.RY(x[2], wires=2)\n",
        "    qml.RY(x[3], wires=3)\n",
        "    qml.RY(x[4], wires=4)\n",
        "    qml.RY(x[5], wires=5)\n",
        "    qml.RY(x[6], wires=6)\n",
        "    qml.RY(x[7], wires=7)\n",
        "    qml.RY(x[8], wires=8)\n",
        "\n",
        "    # Centre single-qubit rotation\n",
        "    qml.RX(p[0], wires=4)\n",
        "    qml.RY(p[1], wires=4)\n",
        "\n",
        "    # Note in this circuit the parameters aren't all the same.\n",
        "    # Previously they were identical to ensure they were applied\n",
        "    # as one combined gate. The fact they can all vary independently\n",
        "    # here means we aren't respecting the symmetry.\n",
        "\n",
        "    # Corner single-qubit rotation\n",
        "    qml.RX(p[2], wires=0)\n",
        "    qml.RX(p[3], wires=2)\n",
        "    qml.RX(p[4], wires=6)\n",
        "    qml.RX(p[5], wires=8)\n",
        "\n",
        "    qml.RY(p[6], wires=0)\n",
        "    qml.RY(p[7], wires=2)\n",
        "    qml.RY(p[8], wires=6)\n",
        "    qml.RY(p[9], wires=8)\n",
        "\n",
        "    # Edge single-qubit rotation\n",
        "    qml.RX(p[10], wires=1)\n",
        "    qml.RX(p[11], wires=3)\n",
        "    qml.RX(p[12], wires=5)\n",
        "    qml.RX(p[13], wires=7)\n",
        "\n",
        "    qml.RY(p[14], wires=1)\n",
        "    qml.RY(p[15], wires=3)\n",
        "    qml.RY(p[16], wires=5)\n",
        "    qml.RY(p[17], wires=7)\n",
        "\n",
        "    # Double Entagling two-qubit gates\n",
        "    # circling the edge of the board\n",
        "    qml.CRY(p[18], wires=[0, 1])\n",
        "    qml.CRY(p[19], wires=[2, 1])\n",
        "    qml.CRY(p[20], wires=[2, 5])\n",
        "    qml.CRY(p[21], wires=[8, 5])\n",
        "    qml.CRY(p[22], wires=[8, 7])\n",
        "    qml.CRY(p[23], wires=[6, 7])\n",
        "    qml.CRY(p[24], wires=[6, 3])\n",
        "    qml.CRY(p[25], wires=[0, 3])\n",
        "\n",
        "    # To the corners from the centre\n",
        "    qml.CRY(p[26], wires=[4, 0])\n",
        "    qml.CRY(p[27], wires=[4, 2])\n",
        "    qml.CRY(p[28], wires=[4, 6])\n",
        "    qml.CRY(p[29], wires=[4, 8])\n",
        "\n",
        "    # To the centre from the edges\n",
        "    qml.CRY(p[30], wires=[1, 4])\n",
        "    qml.CRY(p[31], wires=[3, 4])\n",
        "    qml.CRY(p[32], wires=[5, 4])\n",
        "    qml.CRY(p[33], wires=[7, 4])\n",
        "\n",
        "    # circling the edge of the board\n",
        "    qml.CRY(p[18], wires=[0, 1])\n",
        "    qml.CRY(p[19], wires=[2, 1])\n",
        "    qml.CRY(p[20], wires=[2, 5])\n",
        "    qml.CRY(p[21], wires=[8, 5])\n",
        "    qml.CRY(p[22], wires=[8, 7])\n",
        "    qml.CRY(p[23], wires=[6, 7])\n",
        "    qml.CRY(p[24], wires=[6, 3])\n",
        "    qml.CRY(p[25], wires=[0, 3])\n",
        "\n",
        "    # To the corners from the centre\n",
        "    qml.CRY(p[26], wires=[4, 0])\n",
        "    qml.CRY(p[27], wires=[4, 2])\n",
        "    qml.CRY(p[28], wires=[4, 6])\n",
        "    qml.CRY(p[29], wires=[4, 8])\n",
        "\n",
        "    # To the centre from the edges\n",
        "    qml.CRY(p[30], wires=[1, 4])\n",
        "    qml.CRY(p[31], wires=[3, 4])\n",
        "    qml.CRY(p[32], wires=[5, 4])\n",
        "    qml.CRY(p[33], wires=[7, 4])\n",
        "\n",
        "    return [qml.expval(ob_center), qml.expval(ob_corner), qml.expval(ob_edge)]\n",
        "\n",
        "\n",
        "fig, ax = qml.draw_mpl(circuit_no_sym)([0] * 9, [0] * 34)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1REWjSNvPtCJ"
      },
      "source": [
        "Note again how, though these circuits have a similar form to before,\n",
        "they are parameterised differently. We need to feed the vector\n",
        "$\\boldsymbol{y}$ made up of the expectation value of these three\n",
        "operators into the loss function and use this to update our parameters.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "7wTu3Uw_PtCJ"
      },
      "outputs": [],
      "source": [
        "import math\n",
        "\n",
        "def encode_game(game):\n",
        "    board, res = game\n",
        "    x = board * (2 * math.pi) / 3\n",
        "    if res == 1:\n",
        "        y = [-1, -1, 1]\n",
        "    elif res == -1:\n",
        "        y = [1, -1, -1]\n",
        "    else:\n",
        "        y = [-1, 1, -1]\n",
        "    return x, y"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qMYUsXoZPtCJ"
      },
      "source": [
        "Recall that the loss function we\\'re interested in is\n",
        "$\\mathcal{L}(\\mathcal{D})=\\frac{1}{|\\mathcal{D}|} \\sum_{(\\boldsymbol{g}, \\boldsymbol{y}) \\in \\mathcal{D}}\\|\\hat{\\boldsymbol{y}}(\\boldsymbol{g})-\\boldsymbol{y}\\|_{2}^{2}$.\n",
        "We need to define this and then we can begin our optimisation.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "_RVbTFtXPtCJ"
      },
      "outputs": [],
      "source": [
        "# calculate the mean square error for this classification problem\n",
        "def cost_function(params, input, target):\n",
        "    output = torch.stack([torch.hstack(circuit(x, params)) for x in input])\n",
        "    vec = output - target\n",
        "    sum_sqr = torch.sum(vec * vec, dim=1)\n",
        "    return torch.mean(sum_sqr)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wiKzIjHSPtCJ"
      },
      "source": [
        "Let\\'s now train our symmetry-preserving circuit on the data.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "ISDNjQj0PtCK",
        "outputId": "5e938cbf-2218-46a6-8edb-8ef841c22141"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "accuracy without training = 0.5249999761581421\n",
            "Epoch:  1 | Loss: 3.176805 | Validation accuracy: 0.485000\n",
            "Epoch:  2 | Loss: 2.856527 | Validation accuracy: 0.436667\n",
            "Epoch:  3 | Loss: 2.822027 | Validation accuracy: 0.496667\n",
            "Epoch:  4 | Loss: 2.764042 | Validation accuracy: 0.456667\n",
            "Epoch:  5 | Loss: 2.671184 | Validation accuracy: 0.560000\n",
            "Epoch:  6 | Loss: 2.681827 | Validation accuracy: 0.545000\n",
            "Epoch:  7 | Loss: 2.649758 | Validation accuracy: 0.541667\n",
            "Epoch:  8 | Loss: 2.564894 | Validation accuracy: 0.573333\n",
            "Epoch:  9 | Loss: 2.649880 | Validation accuracy: 0.591667\n",
            "Epoch: 10 | Loss: 2.615056 | Validation accuracy: 0.580000\n",
            "Epoch: 11 | Loss: 2.590838 | Validation accuracy: 0.596667\n",
            "Epoch: 12 | Loss: 2.577285 | Validation accuracy: 0.583333\n",
            "Epoch: 13 | Loss: 2.638542 | Validation accuracy: 0.581667\n",
            "Epoch: 14 | Loss: 2.569011 | Validation accuracy: 0.595000\n",
            "Epoch: 15 | Loss: 2.578797 | Validation accuracy: 0.585000\n"
          ]
        }
      ],
      "source": [
        "from torch import optim\n",
        "import numpy as np\n",
        "\n",
        "params = 0.01 * torch.randn(9)\n",
        "params.requires_grad = True\n",
        "opt = optim.Adam([params], lr=1e-2)\n",
        "\n",
        "\n",
        "max_epoch = 15\n",
        "max_step = 30\n",
        "batch_size = 10\n",
        "\n",
        "encoded_dataset = list(zip(*[encode_game(game) for game in dataset]))\n",
        "encoded_dataset_val = list(zip(*[encode_game(game) for game in dataset_val]))\n",
        "\n",
        "\n",
        "def accuracy(p, x_val, y_val):\n",
        "    with torch.no_grad():\n",
        "        y_val = torch.tensor(y_val)\n",
        "        y_out = torch.stack([torch.hstack(circuit(x, p)) for x in x_val])\n",
        "        acc = torch.sum(torch.argmax(y_out, axis=1) == torch.argmax(y_val, axis=1))\n",
        "        return acc / len(x_val)\n",
        "\n",
        "\n",
        "print(f\"accuracy without training = {accuracy(params, *encoded_dataset_val)}\")\n",
        "\n",
        "x_dataset = torch.stack(encoded_dataset[0])\n",
        "y_dataset = torch.tensor(encoded_dataset[1], requires_grad=False)\n",
        "\n",
        "saved_costs_sym = []\n",
        "saved_accs_sym = []\n",
        "for epoch in range(max_epoch):\n",
        "    rand_idx = torch.randperm(len(x_dataset))\n",
        "    # Shuffled dataset\n",
        "    x_dataset = x_dataset[rand_idx]\n",
        "    y_dataset = y_dataset[rand_idx]\n",
        "\n",
        "    costs = []\n",
        "\n",
        "    for step in range(max_step):\n",
        "        x_batch = x_dataset[step * batch_size : (step + 1) * batch_size]\n",
        "        y_batch = y_dataset[step * batch_size : (step + 1) * batch_size]\n",
        "\n",
        "        def opt_func():\n",
        "            opt.zero_grad()\n",
        "            loss = cost_function(params, x_batch, y_batch)\n",
        "            costs.append(loss.item())\n",
        "            loss.backward()\n",
        "            return loss\n",
        "\n",
        "        opt.step(opt_func)\n",
        "\n",
        "    cost = np.mean(costs)\n",
        "    saved_costs_sym.append(cost)\n",
        "\n",
        "    if (epoch + 1) % 1 == 0:\n",
        "        # Compute validation accuracy\n",
        "        acc_val = accuracy(params, *encoded_dataset_val)\n",
        "        saved_accs_sym.append(acc_val)\n",
        "\n",
        "        res = [epoch + 1, cost, acc_val]\n",
        "        print(\"Epoch: {:2d} | Loss: {:3f} | Validation accuracy: {:3f}\".format(*res))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z9SOd6lQPtCK"
      },
      "source": [
        "Now we train the non-symmetry preserving circuit.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "THRXOBEUPtCK",
        "outputId": "3e4761c0-bdf8-470e-b9d3-eba258c5aac0"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "accuracy without training = 0.44999998807907104\n",
            "Epoch:  1 | Loss: 3.169856 | Validation accuracy: 0.438333\n",
            "Epoch:  2 | Loss: 2.915814 | Validation accuracy: 0.443333\n",
            "Epoch:  3 | Loss: 2.848999 | Validation accuracy: 0.445000\n",
            "Epoch:  4 | Loss: 2.798156 | Validation accuracy: 0.428333\n",
            "Epoch:  5 | Loss: 2.745969 | Validation accuracy: 0.453333\n",
            "Epoch:  6 | Loss: 2.714810 | Validation accuracy: 0.461667\n",
            "Epoch:  7 | Loss: 2.697196 | Validation accuracy: 0.485000\n",
            "Epoch:  8 | Loss: 2.678972 | Validation accuracy: 0.480000\n",
            "Epoch:  9 | Loss: 2.667330 | Validation accuracy: 0.510000\n",
            "Epoch: 10 | Loss: 2.651770 | Validation accuracy: 0.513333\n",
            "Epoch: 11 | Loss: 2.636838 | Validation accuracy: 0.525000\n",
            "Epoch: 12 | Loss: 2.623924 | Validation accuracy: 0.560000\n",
            "Epoch: 13 | Loss: 2.608304 | Validation accuracy: 0.531667\n",
            "Epoch: 14 | Loss: 2.597640 | Validation accuracy: 0.555000\n",
            "Epoch: 15 | Loss: 2.588152 | Validation accuracy: 0.565000\n"
          ]
        }
      ],
      "source": [
        "params = 0.01 * torch.randn(34)\n",
        "params.requires_grad = True\n",
        "opt = optim.Adam([params], lr=1e-2)\n",
        "\n",
        "# calculate mean square error for this classification problem\n",
        "\n",
        "\n",
        "def cost_function_no_sym(params, input, target):\n",
        "    output = torch.stack([torch.hstack(circuit_no_sym(x, params)) for x in input])\n",
        "    vec = output - target\n",
        "    sum_sqr = torch.sum(vec * vec, dim=1)\n",
        "    return torch.mean(sum_sqr)\n",
        "\n",
        "\n",
        "max_epoch = 15\n",
        "max_step = 30\n",
        "batch_size = 15\n",
        "\n",
        "encoded_dataset = list(zip(*[encode_game(game) for game in dataset]))\n",
        "encoded_dataset_val = list(zip(*[encode_game(game) for game in dataset_val]))\n",
        "\n",
        "\n",
        "def accuracy_no_sym(p, x_val, y_val):\n",
        "    with torch.no_grad():\n",
        "        y_val = torch.tensor(y_val)\n",
        "        y_out = torch.stack([torch.hstack(circuit_no_sym(x, p)) for x in x_val])\n",
        "        acc = torch.sum(torch.argmax(y_out, axis=1) == torch.argmax(y_val, axis=1))\n",
        "        return acc / len(x_val)\n",
        "\n",
        "\n",
        "print(f\"accuracy without training = {accuracy_no_sym(params, *encoded_dataset_val)}\")\n",
        "\n",
        "\n",
        "x_dataset = torch.stack(encoded_dataset[0])\n",
        "y_dataset = torch.tensor(encoded_dataset[1], requires_grad=False)\n",
        "\n",
        "saved_costs = []\n",
        "saved_accs = []\n",
        "for epoch in range(max_epoch):\n",
        "    rand_idx = torch.randperm(len(x_dataset))\n",
        "    # Shuffled dataset\n",
        "    x_dataset = x_dataset[rand_idx]\n",
        "    y_dataset = y_dataset[rand_idx]\n",
        "\n",
        "    costs = []\n",
        "\n",
        "    for step in range(max_step):\n",
        "        x_batch = x_dataset[step * batch_size : (step + 1) * batch_size]\n",
        "        y_batch = y_dataset[step * batch_size : (step + 1) * batch_size]\n",
        "\n",
        "        def opt_func():\n",
        "            opt.zero_grad()\n",
        "            loss = cost_function_no_sym(params, x_batch, y_batch)\n",
        "            costs.append(loss.item())\n",
        "            loss.backward()\n",
        "            return loss\n",
        "\n",
        "        opt.step(opt_func)\n",
        "\n",
        "    cost = np.mean(costs)\n",
        "    saved_costs.append(costs)\n",
        "\n",
        "    if (epoch + 1) % 1 == 0:\n",
        "        # Compute validation accuracy\n",
        "        acc_val = accuracy_no_sym(params, *encoded_dataset_val)\n",
        "        saved_accs.append(acc_val)\n",
        "\n",
        "        res = [epoch + 1, cost, acc_val]\n",
        "        print(\"Epoch: {:2d} | Loss: {:3f} | Validation accuracy: {:3f}\".format(*res))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8ujgVimaPtCK"
      },
      "source": [
        "Finally let\\'s plot the results and see how the two training regimes\n",
        "differ.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "GeF6cgOLPtCK",
        "outputId": "9838d950-89b7-44d9-e2ec-f1a6b4ffcd72"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "from matplotlib import pyplot as plt\n",
        "\n",
        "plt.title(\"Validation accuracies\")\n",
        "plt.plot(saved_accs_sym, \"b\", label=\"Symmetric\")\n",
        "plt.plot(saved_accs, \"g\", label=\"Standard\")\n",
        "\n",
        "plt.ylabel(\"Validation accuracy (%)\")\n",
        "plt.xlabel(\"Optimization steps\")\n",
        "plt.legend()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iilNBvUJPtCK"
      },
      "source": [
        "What we can see then is that by paying attention to the symmetries\n",
        "intrinsic to the learning problem and reflecting this in an equivariant\n",
        "gate set we have managed to improve our learning accuracies, while also\n",
        "using fewer parameters. While the symmetry-aware circuit clearly\n",
        "outperforms the naive one, it is notable however that the learning\n",
        "accuracies in both cases are hardly ideal given this is a solved game.\n",
        "So paying attention to symmetries definitely helps, but it also isn\\'t a\n",
        "magic bullet!\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "59M09yH_PtCK"
      },
      "source": [
        "The use of symmetries in both quantum and classical machine learning is\n",
        "a developing field, so we can expect new results to emerge over the\n",
        "coming years. If you want to get involved, the references given below\n",
        "are a great place to start.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AfMCcW_QPtCK"
      },
      "source": [
        "References\n",
        "==========\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_Xne5NFbPtCK"
      },
      "source": [
        "Acknowledgments\n",
        "===============\n",
        "\n",
        "The author would also like to acknowledge the helpful input of C.-Y.\n",
        "Park.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_kP9cOTuPtCK"
      },
      "source": [
        "About the author\n",
        "================\n"
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.9.17"
    },
    "colab": {
      "provenance": []
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}