<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Kathie Wang on Medium]]></title>
        <description><![CDATA[Stories by Kathie Wang on Medium]]></description>
        <link>https://medium.com/@qrst3721?source=rss-274aa921a049------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*i-uCw4Y26UtNv6WH2RpGOw.jpeg</url>
            <title>Stories by Kathie Wang on Medium</title>
            <link>https://medium.com/@qrst3721?source=rss-274aa921a049------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 09:25:06 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@qrst3721/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Quantum Entanglement: The Dance of Our Universe]]></title>
            <link>https://medium.com/@qrst3721/quantum-entanglement-the-cosmic-dance-of-our-universe-dafaf542988b?source=rss-274aa921a049------2</link>
            <guid isPermaLink="false">https://medium.com/p/dafaf542988b</guid>
            <category><![CDATA[entanglement]]></category>
            <category><![CDATA[optics]]></category>
            <category><![CDATA[photonics]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <category><![CDATA[qubit]]></category>
            <dc:creator><![CDATA[Kathie Wang]]></dc:creator>
            <pubDate>Sun, 16 Jun 2024 05:54:09 GMT</pubDate>
            <atom:updated>2024-06-16T13:29:27.420Z</atom:updated>
            <content:encoded><![CDATA[<p>Imagine you are in a dance, spinning like a ballerina or a break dancer. Then, suddenly, as you come to a silent rest, your final position has determined the future of the whole world — and all of the entities of the worlds within the very universe itself.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*ikp0AEm0eQsMJXkqRf9uFA.gif" /><figcaption>Video from Pixabay</figcaption></figure><p>This is what just happened — born from our imagination and our understanding of nature.</p><p>Your life is now completely different, even if you feel exactly the same.</p><p>In May 2024, researchers in the United States, China, and the Netherlands successfully generated quantum entanglement on their own and showed the world that we are getting closer to establishing a “quantum internet!”</p><p>This experiment is a major milestone; in the past, quantum entanglement has been shown in labs, but these research groups produced these entangled photons in a large-scale network setting, which is much more difficult to accomplish.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*TygnnFkrhsVGvyu4ddJ0AQ.gif" /><figcaption>Video from Pixabay</figcaption></figure><p>Before I get into the details of each research group, here are some important terms to understand that are related to the recent experiments:</p><ul><li><strong>Qubit</strong>: the basic unit of quantum information — there are many different types of qubits, such as superconducting, trapped ions, and photonic qubits. In the context of recent discoveries, the lab from China used neutral atoms qubits, and the labs from the United States and the Netherlands used solid-state spin qubits.</li><li><strong>Entanglement</strong>: when two or more objects are linked so that the quantum state of each particle of the group cannot be described independently of the state of the others, even if they are far apart. Quantum entanglement can be generated in many ways, and I will explain the methods of the research groups later in this article.</li><li><strong>Photon</strong>: a tiny particle that is the fundamental unit (smallest particle) of electromagnetic radiation. Photons have no mass and no charge, but have momentum and can travel at the speed of light! In quantum computing, they are used to encode, carry, and manipulate quantum information.</li><li><strong>Quantum internet</strong>: a network that could allow information to be exchanged while encoded in quantum states</li><li><strong>Optical fiber</strong>: a material that transmits signals or photons from one end to the other in the form of light. Optical fibers are typically flexible glass or plastic fibers, and allow for transmission over longer distances than electrical cables.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/943/1*1cwI7tHnQqPHlAlqE1rpww.png" /><figcaption>Image of optical fiber from technologyreview.com</figcaption></figure><p>Now that we reviewed the key terms related to quantum entanglement and the hardware aspect of quantum computing, we can understand what each group did. Each of the three groups stored the quantum information in different ways, but all connected parts of a network using <strong>photons</strong> in the <strong>infrared</strong> part of the spectrum.</p><p>Infrared light is a section of the electromagnetic spectrum with waves slower than those of the visible light we see. There are many reasons why optical fibers are most efficient at transmitting infrared light (both the fibers and the photons play a role), but in summary, data can be transmitted over longer distances with minimal signal degradation.</p><p>At the University of Science and Technology of China (USTC) in Hefei, China, Pan Jian-Wei and his team used a photon to set the quantum states encoded by rubidium atoms — their qubits. To put the nodes in an entangled state, the emitted photons, which carry the quantum information, would have to arrive at their photonic server (which was 10 kilometers away) at the same time. To do this, Pan’s team “tickled” the atomic clouds to emit photons. This means they applied an external field or pulse to control when and how photons are emitted from the excited rubidium atoms.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/805/1*q-EwpQQoQDmL6A9It-EXtQ.png" /><figcaption>Illustration of a rubidium atom producing a stream of entangled photons, from <a href="https://newatlas.com/physics/quantum-entanglement-photons-record/">https://newatlas.com/physics/quantum-entanglement-photons-record/</a></figcaption></figure><p>At the Delft University of Technology in the Netherlands, Ronald Hanson and his research group used qubits that are diamonds with “Nitrogen-Vacancy” (NV) centers. These have a nitrogen atom replacing a carbon atom at one location that is also next to a vacant carbon atom. This allows researchers to use infrared light to manipulate the electron spin states of nitrogen, as well as the nuclear states of neighboring carbon atoms.</p><p>Like Pan’s team, Hanson’s team used a central photonic server to receive the photon signals, and if they arrived at the same exact time, the quantum states of the original qubits could interact in a way that results in entanglement.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/548/1*e54mzhCYdEzj2nhW2U3ZyA.png" /><figcaption>Diamond with an NV center from <a href="https://www.mdpi.com/2504-4494/1/1/6">https://www.mdpi.com/2504-4494/1/1/6</a></figcaption></figure><p>Finally, at Harvard University in the United States, Mikhail Lukin and his team used a diamond with a “Silicon-Vacancy” (SiV) center. This is just like the diamond with the NV center, but it has a missing carbon pair (two missing carbons, not one).</p><p>Lukin’s team first sent one photon with an initial quantum state to the silicon atom, which entangled the photon with the first qubit’s information. When the photon reached the second silicon atom, which was 35 kilometers away from the first, the photon then entangled itself with the second qubit. Now, entanglement has been achieved between two qubits using a single photon!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/513/1*EehzwMBDa1x4YFO-6o6fVw.jpeg" /><figcaption>Diamond with SiV center from <a href="https://www.sciencedirect.com/science/article/pii/S092596352100011X]">https://www.sciencedirect.com/science/article/pii/S092596352100011X]</a></figcaption></figure><p>All the labs created quantum entanglement in different ways, but their overall concept is very similar. To recap, the <strong>qubits</strong> have quantum states that carry quantum information. Then, <strong>photons</strong> facilitate transferring the information and are the entities that can lead to creating entanglement between the two states.<strong> </strong>During this process, <strong>optical fibers </strong>provide an efficient path for the photons to travel in and store information for longer periods.</p><p>These developments have many diverse and critical applications. By demonstrating entanglement over large distances, we can not only effectively protect sensitive information, but also potentially connect separate quantum computers into one larger and more powerful machine, and create networks of high-resolution optical telescopes!</p><p>So as you spin and spin some more, like pieces of clay spinning on a ceramic wheel, you are sculpting the cup and the bowl that will contain, preserve, and spread the elixir of life itself.</p><p>A quantum entangled cosmic dance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*NWnvD1MfLpOJhUdAx5RMqA.gif" /></figure><p>Check out Vidur’s article on the types of qubits and their advantages and disadvantages! <a href="https://medium.com/@vidur.jannapureddy/unveiling-the-quantum-future-a-deep-dive-into-the-world-of-qubits-db165f53c85c">https://medium.com/@vidur.jannapureddy/unveiling-the-quantum-future-a-deep-dive-into-the-world-of-qubits-db165f53c85c</a></p><p>Below are also some sources I used:</p><p><a href="https://www.energy.gov/science/articles/creating-heart-quantum-computer-developing-qubits">https://www.energy.gov/science/articles/creating-heart-quantum-computer-developing-qubits</a></p><p><a href="https://www.nature.com/articles/d41586-024-01445-2">https://www.nature.com/articles/d41586-024-01445-2</a></p><p><a href="https://www.quera.com/neutral-atom-platform">https://www.quera.com/neutral-atom-platform</a></p><p><a href="https://www.forbes.com/sites/chadorzel/2017/02/28/how-do-you-create-quantum-entanglement/">https://www.forbes.com/sites/chadorzel/2017/02/28/how-do-you-create-quantum-entanglement/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dafaf542988b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OMG, I’m starting to understand Quantum Computing!]]></title>
            <link>https://medium.com/@qrst3721/omg-im-starting-to-understand-quantum-computing-3dbae4ee5bad?source=rss-274aa921a049------2</link>
            <guid isPermaLink="false">https://medium.com/p/3dbae4ee5bad</guid>
            <category><![CDATA[bloch-sphere]]></category>
            <category><![CDATA[beginners-guide]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <category><![CDATA[quantum-gate]]></category>
            <category><![CDATA[quantum-computer]]></category>
            <dc:creator><![CDATA[Kathie Wang]]></dc:creator>
            <pubDate>Sat, 11 May 2024 01:35:30 GMT</pubDate>
            <atom:updated>2024-05-19T16:38:19.067Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*kPmXvnnZ4wHLX4YiSUcdYg.gif" /></figure><p>Today, I will be writing about a time when I first realized that I was starting to understand quantum computing and some resources that were very helpful for me!</p><p>Like many others, I first came across Bloch spheres when I started reading about how quantum states are represented on a qubit. Though it made sense to me that the states represented information of a qubit, I struggled with visualizing them since the sphere represents information in 3D, and all I saw were 2D images! I also read how certain gates were “bit flips” and “phase-flips”, and had a hard time keeping track of which gates did what operation. This was until I came across this website: <a href="https://javafxpert.github.io/grok-bloch/">https://javafxpert.github.io/grok-bloch/</a></p><p>Using this website, you can drag the sphere and rotate it any way you want to see the sphere in different perspectives, see the probability the state will be measured to have state 0, and most importantly, experiment and see what different combinations of gates affect the quantum state!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*cx2FypZfLBctX0m78I1zPA.gif" /></figure><h3><em>The Bloch Sphere</em></h3><p>First, let me explain how the Bloch sphere works. The Bloch sphere geometrically represents one qubit, or the state of a qubit. It can either be a pure state — when the vectors end at the surface of the sphere (which has a radius of 1) — or a mixed state — when the vectors stop at the interior. If you are unsure of what a term means, such as pure and mixed states, there will be a vocabulary section below to explain them!</p><p>Since the Bloch sphere represents three dimensions, the X, Y, and Z axes are drawn to show direction, and the center of the sphere is aligned with the origin (0,0,0). The positive end of the X axis is state |+&gt; (pronounced “ket plus”), and the negative end is state |-&gt; (“ket minus”). Similarly, the positive and negative ends of the Y and Z axes are |i&gt;, |-i&gt; and |0&gt;, |1&gt; respectively.</p><p>The Z-axis represents the probability of the qubit being measured as a 0 or a 1, the X-axis represents the real part of the state vector, and the Y-axis represents its imaginary part. Oftentimes we use the Z basis when defining states because quantum computers perform all measurements in the Z basis (also called the standard computational basis).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*kXNYSv32PpSgku2MtrjACA.png" /><figcaption>Image from <a href="https://www.researchgate.net/figure/color-online-Bloch-sphere-representation-of-the-Bloch-vector-v-for-qubits-d-2_fig3_316828699">researchgate.net</a></figcaption></figure><h4>Examples</h4><p>Here are four examples that I find are the most interesting and helped me visualize the purpose of quantum operators the most:</p><ol><li><strong>Pauli Gates</strong></li></ol><p>Using the website from above, you can see that there are three gate icons in light green. These are known as the Pauli X, Y, and Z gates, and just like all the other gates, they are fundamental to manipulating quantum states. The Pauli X gate is known as a “bit flip” gate because it flips a state 0 to state 1, and vice versa. This is possible because quantum states and operators are represented by matrices, and you can determine the outcome of applying a gate/operator to a state by multiplying the matrices.</p><p>You can also think of the X gate as an operator that rotates the current state by π radians of 180 degrees around the X axis. This is tricky to visualize with just a 2D image, so this gif shows what the rotation looks like.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/360/1*sPISGzEFC0e6SPLy5RCCqg.gif" /><figcaption>Original source from <a href="https://twitter.com/j_bertolotti/status/1142503314287943680">https://twitter.com/j_bertolotti/status/1142503314287943680</a></figcaption></figure><p>Now that you understand the X gate, the Y and Z gates will also make more sense! All the Pauli gates perform a π radian rotation about their respective axis, and clicking the same gate twice brings you back to its initial state. Something to note is that the Pauli Z gate is called a “phase-flip” because it keeps the probabilities of measuring 0 or 1 the same, but swaps the probabilities of the qubit being + or -.</p><p><strong>2. The Hadamard Gate</strong></p><p>Now, if you want to “reset” your sphere, you can also set the qubit to state |0&gt;. Then, try clicking on the H gate, which is below the Pauli Z gate. You should see that unlike the Pauli gates that perform a rotation by π radians, it rotates the state by rotation states |0&gt; and |1&gt; to |+&gt; and |+&gt; respectively. Feel free to verify this by applying the H gate after different initial states!</p><p>As I mentioned before in previous articles, applying the Hadamard gate puts the state into an equal superposition of the |0&gt; and |1&gt; states. This means that there is an equal probability of measuring the states 0 and 1 — each probability is 1/2. This is shown by the bar on the left side of the Bloch sphere screen, which displays the probability of measuring |0&gt;. When you apply the H gate with the initial state as |0&gt;, the bar changes from a probability of 1 to 0.5, as indicated by the dark “shaded” rectangle. You can also check out the mathematical proof <a href="https://quantumcomputinguk.org/tutorials/hadamardgate?rq=hadamard">here</a>.</p><p><strong>3. Phase Gates</strong></p><p>Two more phase gates are the T and S gates. The T gate performs a phase shift of π/4 radian rotation about the Z axis, and the S gate a phase shift of π/2 to the |1&gt; state. There are also Rx, Ry, and Rz gates that are orange. You can also apply Rz (rotation about the Z axis) of θ=π/8 twice to get the same outcome of a T gate. There is also an inverse T and inverse S gate, which is an Rz by -π/4 and -π/2 respectively.</p><p>If your sphere is set at state 0, you probably won’t see anything since nothing happens when you are performing a rotation on the Z axis, and the vector is already on the axis. When at state 0, you can then apply the H gate so it’s not on the Z axis to see what the T and S gates do.</p><p><strong>4. Combinations of Gates</strong></p><p>Now, you can experiment with different combinations of the gates. Some I want to highlight are that applying the order of HZH is the same as an X gate (bit-flip), and the HXH operator is a phase-flip gate since it’s the same as the Z gate. Also, applying two T gates in a row is the same as an S gate, which makes sense since π/4 + π/4 = π/2.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*7fePsjP7eV-Iklp3GNAjjg.gif" /></figure><h3><strong>Summary</strong></h3><p>Today, you learned about how the Bloch sphere works, and some of the basic quantum operators. Some terms may be new to you, so here is a list of key words I think would help you better understand quantum computing.</p><p><strong>Pure state: </strong>a quantum state whose state can be defined by a single state vector</p><p><strong>Mixed state: </strong>a quantum state whose state <em>cannot</em> be defined by a single state vector. It is instead defined by a density matrix, and it is essential to representing concepts related to quantum entanglement.</p><p><strong>Operator</strong>: mathematical constructs that can be represented as a matrix, and transforms an input matrix</p><p><strong>Gate</strong>: a specific type of operator used in quantum computing that manipulates qubits in a reversible way</p><p>Recognizing how these gates affect the final state of a qubit was a key moment for me, so I hope this article helps you understand the Bloch sphere and quantum computing better!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*_URh303G7_Xz_efVB3769Q.gif" /></figure><p>Make sure to check out Vidur’s article <a href="https://medium.com/@vidur.jannapureddy/cracking-the-quantum-code-my-eureka-moment-with-the-unfathomable-5599ce74394c">here</a> on his “OMG” moment!</p><p>Below are some links for further reading:</p><ul><li><a href="https://learning.quantum.ibm.com/tutorial/explore-gates-and-circuits-with-the-quantum-composer#hadamard-gate">https://learning.quantum.ibm.com/tutorial/explore-gates-and-circuits-with-the-quantum-composer#hadamard-gate</a></li><li><a href="https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.SGate">https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.SGate</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3dbae4ee5bad" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Darkwing Duck & Quantum Classification]]></title>
            <link>https://medium.com/@qrst3721/darkwing-duck-quantum-classification-741d8afc759d?source=rss-274aa921a049------2</link>
            <guid isPermaLink="false">https://medium.com/p/741d8afc759d</guid>
            <category><![CDATA[classification]]></category>
            <category><![CDATA[quantum-computer]]></category>
            <category><![CDATA[support-vector-machine]]></category>
            <category><![CDATA[quantum-machine-learning]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <dc:creator><![CDATA[Kathie Wang]]></dc:creator>
            <pubDate>Fri, 26 Apr 2024 22:54:47 GMT</pubDate>
            <atom:updated>2024-04-27T12:55:43.600Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Beginner’s Guide to Quantum Support Vector Machines</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*xy_yAjGbIDmXrBtQJ7CBtQ.gif" /></figure><p>Recently, we find ourselves lightning-struck with the terms “machine learning,” “classification,” and “supervised”. But many of us don’t know how these terms are related! This can all become quickly jumbled up in our minds and hinder us from fully understanding how these powerful algorithms were designed, developed, and constructed! This can charr us into a husk of misunderstanding and confusion leading to non-understanding.</p><p>For me, I initially thought that the three terms were on the same “level” to describe an algorithm, that a model is either a machine learning or classification algorithm. This especially confused me when I read about studies using classification <strong>and</strong> supervised machine learning! It turns out machine learning is the general term, and classification is only one big category under machine learning. For <strong>supervised classification</strong>, the process can be broken down into two parts: the learning step and the classification step.</p><p>In the learning step or training phase, the model is trained using the <strong>training data</strong>, a dataset that teaches what the expected output should look like. The training data contains the input objects and each of their desired output values, which allows the classification algorithms to “learn” what characteristics are associated with what labels and outputs. By learning the relationships and “rules”, the classification algorithm can classify similar data points correctly. Next, in the classification step, the model evaluates new unseen <strong>test data</strong> and performs predictions based on what it has previously learned.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CvsxL0_fDzEKE9qlheyC-w.jpeg" /><figcaption>image from labellerr.com</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yOwydMTgik-onxaUgR00Ow.png" /><figcaption>image from analytixlabs</figcaption></figure><p>There are even more subcategories within classification algorithms! Each classification algorithm has a similar goal of sorting the new data into its correct group, but reaches its conclusion through different methods. Today, I will be focusing on specifically the concept of <strong>support vector machines</strong>, and how the quantum version differs from its classical counterpart.</p><p><strong>Important Vocabulary</strong></p><p>To explain how quantum support vector machines (<strong>QSVMs</strong>) work, I have compiled below definitions of important terms used when describing SVMs. Take your time when reading these <em>in order</em> and really try to visualize the descriptions!</p><p><strong>Hyperplane</strong>: a boundary that separates different classes of data points. When the data points are on a high dimension, the plane/boundary will also be a high dimension, so it’s called a hyperplane. Just like how a 2D plane (a standard graph) is split by a 1D plane (a line), a 5-dimensional plane can be split by a 4-dimensional plane. For QSVMs, hybrid classical-quantum optimization techniques are used to find the hyperplane.</p><p><em>Importance</em>: Finding the best hyperplane is the main goal of SVMs because the hyperplane decides how test data points are categorized!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rTisb0DIx_yrlQSprja1TQ.png" /><figcaption>This image from pennylane shows how the original data can’t be separated with a line, but when transformed to a higher dimension, the red and blue data can be split by a hyperplane, which looks like a line!</figcaption></figure><p><strong>Feature map: </strong>a map that transforms the raw features of a dataset by mapping the data points that are in a lower-dimensional feature space onto a high-dimensional space. Quantum feature maps, in particular, use quantum gates and circuits to transform classical data (typically in vector form) into quantum states, which are represented as qubits.</p><p><em>Importance</em>: With the data in a higher dimension, the algorithm can capture more complex patterns in the data. Feature maps also allow SVMs to perform linear classification in a higher-dimensional space, even when the original data is not linearly separable in its original space.</p><p><strong>Margin</strong>: the nearest data points of each class to the selected hyperplane. In SVMs, the optimal hyperplane is the one that splits the data best AND has the greatest margin.</p><p><em>Importance</em>: By maximizing the margin subject to the constraints of the optimization problem, we are closer to obtaining a model with good generalization ability.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/450/1*x5uyav9j_bqT2SIFz-0kKQ.gif" /></figure><p><strong>Kernel function:</strong> a function that uses the high-dimensional data and computes the similarity between two points in the feature map. In SVMs, the “similarity” of two points is the inner product between the points, and in QSVMs, this is accomplished through quantum circuits and algorithms.</p><p><em>Importance</em>: The kernel function incorporates all the vocab terms above together — it decides which hyperplane has the greatest margin and best separates the data points on the feature map, allowing the model to make predictions with high accuracy.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/550/1*MuJ-PhtVh4twE2uD_QVgmA.jpeg" /><figcaption>This <a href="https://www.mdpi.com/1424-8220/19/23/5219">image</a> shows a (n-1)-dimensional hyperplane of data mapped to the nth dimension (the “support vectors” are data points that influence the hyperplane). There are many possible hyperplanes to categorize the data, but the plane with the largest margin is the best one.</figcaption></figure><p><strong>Putting It All Together</strong></p><p>To create the quantum support vector machine, the first step is to encode the classical data into quantum states using a <strong>quantum feature map.</strong> In the quantum circuit, the parameters of the quantum gates encode information about the classical data. The quantum circuits are then used to define the <strong>kernel function</strong>, by applying quantum gates to manipulate the encoded states described in, and then measuring the quantum states to compute the inner product between them.</p><p>Next is the training phase. The QSVM model is trained using the learning/training data, finding the <strong>best hyperplane</strong> that separates the data points in the high-dimensional space, or the plane that <em>maximizes</em> the <strong>margin</strong>. Finally, during the testing phase, new unlabeled data is classified according to solution found in training phase.</p><p>In short, QSVMs classify data points into different classes by finding the optimal hyperplane that maximizes the margin between the classes, so unseen data can be correctly classified in the future. They use quantum computers to prepare the quantum states and evaluate the kernel function, and use classical computers to perform the optimization required to find the optimal hyperplane.</p><p><strong>How does a quantum SVM differ from a classical SVM?</strong></p><p>While classical support vector machines are already very well established, their quantum versions can be even more powerful! Below are the three main advantages of QSVMs:</p><ol><li>Higher-Dimensional Mapping: When encoding the data to the higher-dimension feature maps, quantum computers can compute multiple states at the same time. This can lead to better separation of data points in complex feature spaces compared to classical kernels. Also when finding the kernel function, quantum computers calculate the inner products faster than their classical counterparts because they utilize the quantum phenomena of superposition and entanglement.</li><li>Faster training times: Due to increased computational power of quantum computers, the optimal hyperplane can be found faster than the process for classical computers. This can be especially useful for large-scale machine learning tasks where training time can be an issue.</li><li>Better generalization performance: Since quantum computers use quantum mechanics (and properties including superposition and entanglement) to represent the data, the data are represented in a more efficient way. QSVMs have actually been shown to make more accurate predictions on unseen data than classical SVMs.</li></ol><p><strong>Summary</strong></p><p>Today, you learned about quantum support vector machines, a type of classification algorithm in supervised machine learning. You learned about how quantum feature maps, kernel functions, margins, and hyperplanes all play a role in training the model, so it can make accurate predictions on future data. I hope this article cleared up some confusion on the key words in quantum machine learning, and that you understand SVMs better!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*xvPOGraHTVq0xM_aGqEeCQ.gif" /></figure><p><a href="https://medium.com/@vidur.jannapureddy/unleashing-the-quantum-pickle-rick-a-dive-into-quantum-neural-networks-and-how-they-could-get-you-36a95b0f406c">Click here</a> to check out Vidur’s article on neural networks!</p><p><strong>References</strong></p><p><a href="https://www.mdpi.com/1424-8220/19/23/5219">https://www.mdpi.com/1424-8220/19/23/5219</a></p><p><a href="https://www.geeksforgeeks.org/major-kernel-functions-in-support-vector-machine-svm/">https://www.geeksforgeeks.org/major-kernel-functions-in-support-vector-machine-svm/</a></p><p><a href="https://dzone.com/articles/quantum-support-vector-machine-101">https://dzone.com/articles/quantum-support-vector-machine-101</a></p><p><a href="https://www.analytixlabs.co.in/blog/classification-in-machine-learning/">https://www.analytixlabs.co.in/blog/classification-in-machine-learning/</a></p><p><a href="https://www.labellerr.com/blog/supervised-vs-unsupervised-learning-whats-the-difference/">https://www.labellerr.com/blog/supervised-vs-unsupervised-learning-whats-the-difference/</a></p><p><a href="https://pennylane.ai/qml/demos/tutorial_kernels_module/">https://pennylane.ai/qml/demos/tutorial_kernels_module/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=741d8afc759d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Quantum Earthquake]]></title>
            <link>https://medium.com/@qrst3721/the-quantum-earthquake-380e476793eb?source=rss-274aa921a049------2</link>
            <guid isPermaLink="false">https://medium.com/p/380e476793eb</guid>
            <category><![CDATA[qubit]]></category>
            <category><![CDATA[quantum-algorithms]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <category><![CDATA[classical-computing]]></category>
            <category><![CDATA[hybrid-quantum-computing]]></category>
            <dc:creator><![CDATA[Kathie Wang]]></dc:creator>
            <pubDate>Sat, 06 Apr 2024 04:11:48 GMT</pubDate>
            <atom:updated>2024-04-13T16:33:32.535Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>What are Quantum Algorithms and How do they Work?</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lCA_DlmWNx5Mxrs_0alsmA.jpeg" /><figcaption>Image from iStock</figcaption></figure><p>When you start your quantum journey, you might feel like you are sitting on an earthquake! As I write this I really am, we had an earthquake here in the Northeastern United States! Even a small one can be scary — and so can quantum algorithms! — so today I want to alleviate the fear and shock of a quantum earthquake. In this article, I will explain how quantum algorithms work, how they differ from their classical counterparts, and how you can go from understanding classical algorithms to quantum algorithms.</p><p>Like many others, I first learned about classical algorithms, including the bubble sort, selection sort, and binary search algorithms to name a few. For me, it helped me better understand the basics of quantum algorithms since both types of programs are categorized in similar ways, such as by their function (ex. sorting or searching). Although other properties of classical algorithms, including time complexity, are not focused as much for quantum algorithms, one way we can categorize quantum algorithms is by the way they are executed. The two categories I’ll be explaining today are pure and hybrid quantum algorithms.</p><p>To provide some background information, quantum computers are made up of quantum bits or qubits, which utilize the “strange” principles of quantum mechanics. One such property is superposition, and in simple terms, it means that particles can exist in a state where they can be in multiple states at the same time! While classical bits exist as either 0 or 1, a qubit can represent information by existing as a combination of states 0 and 1. This allows quantum computers to perform certain calculations very quickly, as well as output a superposition of values.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YzXTss9ipuTJoWpz-GNurA.jpeg" /><figcaption>Image from Center for Quantum Nanoscience, Episode 4</figcaption></figure><p>Another property that is used all the time is entanglement. Entanglement is when a pair or group of particles is connected such that the quantum state of each particle can’t be described independently of the other particles’ states. Instead of manipulating each bit individually like in classical computing, quantum computers can manipulate many qubits at once, suggesting that some problems can be completed more speedily with entangled qubits. Combining superposition and entanglement, we see that quantum computers can process and calculate data much faster than classical computers, especially when the data set is very large.</p><p>Now, pure quantum algorithms are programs that are executed completely on quantum computers. As mentioned earlier, qubits can represent multiple states simultaneously because of their property of superposition. In Qiskit, you can simulate doing this by adding a Hadamard (H) gate to a quantum circuit. As you read more Qiskit tutorials, you will see that the H gate is unique to quantum computing, and it is a helpful way to use the properties of quantum mechanics computationally.</p><p>Meanwhile, for real physical quantum computers, scientists “add a Hadamard gate” by using microwave beams and bouncing them off the qubits. This requires a lot of precision, so even if there was the slightest electromagnetic disturbance, a quantum computer’s calculations would be affected.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*eI-HEUAhSl-iu10-FlL86A.png" /><figcaption>A Hadamard Gate puts a qubit into superposition, where it has a 1/2 probability of being measured state 0 <strong>and</strong> 1 (https://shorturl.at/kL036)</figcaption></figure><p>As you can see, it seems quite difficult to only use quantum computers to execute the quantum algorithms. Hybrid quantum algorithms, on the other hand, use both classical and quantum computing resources. These programs are also known as variational quantum algorithms or VQAs, and they use a <strong>classical</strong> optimizer to train the parameters of a state prepared by a <strong>quantum</strong> circuit. Another important component of VQAs is cost functions, or loss/objective functions. This term pops up everywhere in hybrid quantum algorithms because cost functions reveal if an algorithm is making many errors or not. In other words, VQAs aim to reduce their errors by minimizing their cost function.</p><p>Furthermore, while cost functions are used in classical machine learning, the cost functions in VQAs are evaluated on a quantum computer, which can be more efficient than their classical counterparts thanks to superposition again! By combining specific parts of classical and quantum computing, we can create powerful algorithms that perform complex optimization tasks more quickly and cost-effectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GpWbC75pMR7w29QiPL8S_Q.png" /><figcaption><strong>Schematic of a hybrid quantum algorithm</strong> (learning.quantum.ibm.com/course/variational-algorithm-design)</figcaption></figure><p><strong>Recap</strong></p><p>In this article, we went over the following quantum computing concepts:</p><ul><li>Classical vs Quantum Algorithms</li><li>Superposition</li><li>Entanglement</li><li>Quantum vs Classical Bits</li><li>Pure Quantum Algorithms</li><li>Hybrid/Variational Quantum Algorithms</li><li>Cost Functions</li></ul><p>I would like to end with an important point to keep in mind — quantum algorithms and computers are <em>not</em> meant to replace classical computers, but they may be more efficient in certain topics, such as factorization, classification, and many more! As seen with VQAs, combining both quantum computing and classical machine learning can be more powerful than their separate implementations. I hope this has helped clear up any questions you had about quantum algorithms!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AYyGDSrtVX3xZ29ggGBuLA.jpeg" /></figure><p>Make sure to check out Vidur’s article on specific examples of quantum algorithms and how they work!</p><p><a href="https://medium.com/@vidur.jannapureddy/making-a-quantum-brain-chomp-your-way-through-complex-problems-8b5510bcaaed">https://medium.com/@vidur.jannapureddy/making-a-quantum-brain-chomp-your-way-through-complex-problems-8b5510bcaaed</a></p><p><em>For some more readings, check out these links below:</em></p><ul><li><a href="https://www.quantum-inspire.com/kbase/superposition-and-entanglement/">https://www.quantum-inspire.com/kbase/superposition-and-entanglement/</a></li><li><a href="https://www.quera.com/glossary/noise">https://www.quera.com/glossary/noise</a></li><li><a href="https://ionq.com/resources/what-is-hybrid-quantum-computing">https://ionq.com/resources/what-is-hybrid-quantum-computing</a></li><li><a href="https://learning.quantum.ibm.com/course/variational-algorithm-design/cost-functions">https://learning.quantum.ibm.com/course/variational-algorithm-design/cost-functions</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=380e476793eb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction to Quantum Computing — Start Your First Project!]]></title>
            <link>https://medium.com/@qrst3721/introduction-to-quantum-computing-start-your-first-project-fa42cfb86c7c?source=rss-274aa921a049------2</link>
            <guid isPermaLink="false">https://medium.com/p/fa42cfb86c7c</guid>
            <category><![CDATA[beginners-guide]]></category>
            <category><![CDATA[quantum-computer]]></category>
            <category><![CDATA[jupyter-notebook]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <dc:creator><![CDATA[Kathie Wang]]></dc:creator>
            <pubDate>Sat, 23 Mar 2024 01:24:38 GMT</pubDate>
            <atom:updated>2026-06-15T19:28:53.784Z</atom:updated>
            <content:encoded><![CDATA[<h3>Quantum Pathfinder: Blast Through Beginning Roadblocks &amp; Join the Revolution! Part 2</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Wc05igYH56zVIojpUPyZgg.jpeg" /></figure><p>Hi my name is Kathie, and I have been exploring quantum algorithms and using Qiskit for close to two years. I have my own independent project about predicting mental health status of individuals using Quantum Support Vector Machines, and have been an intern at research groups exploring quantum machine learning and hardware.</p><p>Today I will be answering two main questions: 1) how do I run my code on a Jupyter Notebook, and 2) what is a simple quantum computing project I can do using Qiskit?</p><p>Once you have installed and opened the Jupyter Notebook, you can first practice getting used to the notebook by coding a non-quantum program to reinforce your understanding of Python data types.</p><p>For this example, we can create a dictionary, and print each element while traversing through it (traversing means going through the dictionary from beginning to end). First, we can create a dictionary object and assign the variable name my_dict to it. We can then initialize our dictionary by setting key value pairs to my_dict as shown below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/998/1*p9YdEGQsnPIWC4myCmA-Iw.png" /></figure><p>Keys and values are what makes dictionaries unique, because you can access the corresponding <em>value</em> of a <em>key</em> by just knowing the key. To print all the keys of my_dict, we can write and run the following (run this cell by either clicking the run button on top, or using the shortcut S<em>hift + Enter).</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/496/1*uRI015H8UXXrvuY0-1C8Yg.png" /></figure><p>We can also access only the keys and values of a dictionary using the keys() and values() methods. Before running our program, we should also make sure to periodically save our code to make sure we don’t lose our progress.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/502/1*PdWJJ-NkPa0vmeLL5rXvTQ.png" /></figure><p>Being able to iterate through a dictionary can be very important when you are observing the results of quantum algorithm outputs because you may only need to retrieve one element of the dictionary, and you can do this by traversing through a dictionary.</p><p>Now, we can try writing our first quantum program — constructing a quantum circuit and measuring the counts! Though you may have seen similar tutorials before, I will include common issues you might encounter and how to fix them.</p><p>In a new cell, we will need to import the necessary libraries:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Obkf5NjoX0A9VOAO08wsug.png" /></figure><p>If you run this cell and get an error saying something similar to “cannot import name Execute from ‘qiskit’ ”, this means that the name was typed incorrectly. In this case, you can see that we were trying to import Execute; however, the correct name is the word in lowercase, execute. It’s important to remember that Python is a case-sensitive language, meaning that uppercase and lowercase words represent different things.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/659/1*65lVpHB_sTRJp4CKqsxEDQ.png" /><figcaption>The bottom cell shows the incorrect import from qiskit, while the top cell shows the correct import names</figcaption></figure><p>Next, we can construct a quantum circuit qc by creating two quantum registers (the first number in the parenthesis) and two classical registers (the second number in the parenthesis). To add gates or operators to our circuit, we can use the format qc.[gate name] — this example shows adding a Hadamard gate to quantum register or qubit 0, and a CNOT gate, where qubit 0 is the control qubit and qubit 1 is the target qubit.</p><p>We can then measure the qubits in our circuit by measuring qubit 0 to classical register 0, and qubit 1 to classical register 1. Finally, we can print our circuit by using the qc.draw() method, and include <em>“mpl”</em> and <em>style=“clifford”</em> as a parameters to generate a colorful circuit:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/382/1*Nhp2n7AFSKWZOaJmERA0Gg.png" /></figure><p>Now, we can simulate running our circuit on a real device and get the results. In a new cell, we can create a backend to run our quantum circuit on. We can then use the execute() method to return a job, which keeps track of the results of our experiment. Lastly, we can get the results of our experiment, and get the counts by using their respective methods. We then print the counts, and after saving our file and running all cells, we get the counts display, and see that its data type is a dictionary!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/406/1*Qb0EastHAc8UXJbemmhbUQ.png" /></figure><p>From the printout, we see that the outcome 00 was measured 505 out of 1024 times, and 11 was measured 519 out of 1024 times.</p><p>If you can’t see your circuit drawing, it may be because you wrote the code for the circuit simulation in the same cell as drawing the circuit. You can either add the end section of code in another cell and run both cells, or change the qc.draw() line to display(qc.draw()).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/545/1*pspnwukUdJCDpPCpddHEFA.png" /><figcaption>Line 8 can be changed to display(qc.draw()) if you want to keep your code all in one cell</figcaption></figure><p>It’s also important to run your cells in the correct order, because if a variable is updated in another cell, you will have to run all the other cells that refer to the variable to get the accurate output.</p><p>When I first started learning Qiskit on my own, something that was frustrating for me was forgetting to run the blocks in the correct order. I wanted to separate some lines of code from others to see how each line worked, but it quickly became inconvenient and overwhelming. I’m here to tell you, <strong>don’t worry, you will figure this out!!</strong> An important skill quantum programmers have is being able to carefully examine your own work and see what may be the bug; so remember to not get frustrated when something does not work. Another thing to note is that sometimes you just need to step away from your code for 5 minutes and shift your focus, because you might see your code in a different perspective later. The more projects you do, the more comfortable you will become with programming using Qiskit!</p><p>Check out this link about accessing the IBM quantum lab, downloading your Jupyter notebook, and how to use the lab environment!</p><p><a href="https://medium.com/@vidur.jannapureddy/quantum-pathfinder-blast-through-beginning-roadblocks-join-the-revolution-part-1-19db051acf85">https://medium.com/@vidur.jannapureddy/quantum-pathfinder-blast-through-beginning-roadblocks-join-the-revolution-part-1-19db051acf85</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fa42cfb86c7c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>