QCourse551-1 “Quantum Software Development with Classiq” is free & fully-virtual running between Sep 26, 2023 to Jan 23, 2024. It is a Pass/Fail course, and the study load is equivalent to 6 ECTS credits (around 6-8 hours per week). We will link QCourse551-1 to an actual course at DF@LU (more details will be shared later).
QCourse551-1 provides industrial project development experiences in an academic environment. The Classiq Team offers 13
14 projects as well as mentoring using the Classiq platform. All projects are open-sourced. We have admitted 129 will admit around 80 knowledgeable students. The applications were collected by will be invitation based. Any person providing misleading or wrong information on her background or knowledge will be removed from the QCourse551-1 immediately!
Tentatively, each project will have two groups, and each group will have three members. Each group is expected to report in written form weekly and present their progress in every other week during the online lecture sessions. At the end, each group is asked to prepare a manuscript up to 4 pages, and then we complete the semester with their final presentation. Each activity will be graded by mentors & QCourse guides.
The Classiq slack channel will be the base platform for QCourse. All communications & announcements will be made there. The online sessions will be conducted via zoom meetings. For project management, we will use a dedicated GitLab repository. We will create an issue for each project, and the all progress reports will be weekly posted under the related issue. The project can have its own repo, decided by the mentor.
More details & the list of projects are given below.
Contact: qclass [at] qworld.net
The online classes will start at 17:00 (Riga) on Tuesdays from 90 to 120 minutes. Weekly schedule is below.
|1||Sep 26||Introduction & group meetings|
|2||Oct 3||CLASSIQ Bootcamp|
|3-4||Oct 10, Oct 17||The first progress reports and their presentations|
|5-6||Oct 24, Oct 31||The second progress reports and their presentations|
|7-8||Nov 7, Nov 14||The third progress reports and their presentations|
|9-10||Nov 21, Nov 28||The fourth progress reports and their presentations|
|11-12||Dec 5, Dec 12||The fifth progress reports and their presentations|
|13||Dec 19||Completing the missing reports|
|14-15||Jan 2, Jan 16||Paper writing and its presentations|
|16||Jan 23||Final presentations|
Each project has a subset of following prerequisites:
- [Python] Experience with Python
- [QAlgo] Basic knowledge of quantum algorithms
- [Math] Strong mathematical background
- [Search] Basic understanding of data structures and classical search algorithms
- [QM] Good understanding of quantum mechanics including quantum dynamics
- [QSim] Familiarity with the concept of quantum simulation
- [Edu] Good educational capabilities
- [NN] Basic knowledge in neural networks
- [Signal] Basic knowledge of signal processing
- [QLogic] Basic knowledge of quantum logic and quantum computation
- [ProgLang] Basic understanding of programming languages and compilation
-  Integration of Mitiq with Classiq
prerequisites: [Python] [QAlgo]
The Classiq platform enables generating optimized quantum circuits and to execute them on any quantum computer available on the cloud. In order to gain the most of today’s hardware, error mitigation techniques have been devised. These are applied on the quantum circuit themselves.
The project goal is to incorporate error mitigation on the optimized circuits generated by the Classiq platform using the mitiq package. That is, to incorporate the error mitigation into the flow:functional model -> compiling a quantum circuit with Classiq -> adapting circuit with error mitigation techniques with Mitiq -> executing on quantum hardware using Classiq
Link: Mitiq package
-  Sparse state preparation implementation
prerequisites: [Math] [Python]
State preparation is a key building block in quantum computing. It takes a classical probability vector and generates a quantum state that corresponds to this probability vector. It is common that this probability vector is sparse, i.e. that the number of non-zero probabilities is small.
The goal of this project is to implement a function with the Classiq SDK package that implements an efficient sparse state preparation. The algorithm is based on this paper (and is explained in this video)
Links: Paper & Video
-  Quantum search for bitcoin mining
prerequisites: [Search] [QAlgo]
Bitcoin mining is a perfect example for a use case of the famous quantum search algorithm. In this project you will implement a small example of bitcoin mining using the Classiq platform, and perform a resource estimation of what is needed for doing this in practice. This blog is a good place to start reading on the project.
Link: Blog post
-  Quantum utility with Classiq
prerequisites: [QM] [QSim] [QAlgo]
IBM published lately the paper ‘Evidence for the utility of quantum computing before fault tolerance’. In this paper they executed quantum simulation of a specific Hamiltonian on a quantum hardware. They chose the Hamiltonian by hand to suit perfectly the connectivity map of their hardware.
One of the best advantages of the Classiq platform is it can adapt the algorithm according to the hardware connectivity map automatically! In this work you’ll use this key feature and implement a quantum simulation use case adapted for different hardwares.
-  Hamiltonian simulation application with 25 qubits and depth of 1000
prerequisites: [QM] [QSim] [QAlgo]
Hamiltonian simulation will argubaly be the first algortihm that will exhibit quantum advantage. In order to probe what’s the current status of such an application, we want to run it on real hardware availble nowadays that gives reasonable results. Typical trapped Ions machines availble today are able to run computations of 25 qubits with a circuit depth of up tp 1000 while returning meaningful results.
The goal of this project is to find a specific use case for Hamiltonian simulation and using Classiq, to generate a circuit that implements it with 25 qubits and a circuit depth smaller than 1000.
-  Designing interactive introduction to quantum computing course with Classiq
prerequisites: [Edu] [QAlgo]
Introduction to quantum algortihms could be intimidating and challenging. However, using an intuitive API with state-of-the-art visualization techniques could enable newcomers to get into the field quickly, with good understanding.
The goal of this project is to devise the content that will take people with no quantum algorithms background to understanding the basics algorithms using Classiq API and visualizer. The content should include: state preperation, Grover’s algortihm, Simon algorithm, quantum Fourier transform and quantum phase estimation.
-  QNN algorithm Implementation
prerequisites: [NN] [QAlgo]
Classical neural network has brought advanced capabilities to solve problems we couldn’t before.
Quantum neural networks involves combining classical neural networks with the advantage of quantum information to create more efficient algorithms.
The goal of this project is to understand how we can utilize the quantum capabilities to create quantum layers.
We will create a hybrid network consists of both classical and quantum layers to classify the MNIST dataset.
-  Signal processing primitives with quantum algorithms
prerequisites: [Signal] [QAlgo]
Discrete siganl processing (DSP) has revolutionize our world with applications in everything we encounter – computers, communications, medical devices, transportation etc. Quantum computers hold the promise to disrupt the signal processing realm. In this project, you will invent and design quantum algorithms that implement the basic building block required for DSP, specifically: low pass filters (LPF), band pass filters (BPF), high-pass filters (HPF), expanders and decimeters.
Link: MIT DSP course
-  Maximizing the QAOA
prerequisites: [Python] [QAlgo]
The quantum approximate optimization algorithm (QAOA) is one of the most promising quantum algorithms to reach the first practical quantum advantage. However, the number of iterations needed to get a good solution for a combinatorial optimization problem is problematic for the actual use of this algorithm.
The goal of this project is to get to know several methods to increase the probability of measuring a good solution with a minimum number of iterations. As a project extension, we can also learn how to put constraints for combinatorial optimization problems that are very common in the industry and study how to tune the relevant parameters.
Links: Paper 1 & Paper 2
-  Hybrid HHL
prerequisites: [Python] [QLogic]
The HHL (Harrow, Hassidim, Lloyd) quantum algorithm— for solving a set of linear equations— promises an exponential speedup over its classical counterpart. Typically, implementations of the HHL algorithm involve deep circuits with many qubits, making it inapplicable for near-term devices. One of the approaches to reduce circuit’s depth is to use an hybrid quantum-quantum algorithm, where a Quantum Phase Estimation Algorithm feeds classical information into a reduced HHL algorithm.
The goal of this project is to model and explore the implementation of such hybrid approach, studying relations between circuit depth, width, and functional error. Using the Classiq platform will allow flexiable functional modeling, further gate-count optimization, and examining large use-cases.
-  Quantum Backtracking Implementation
prerequisites: [Search] [QAlgo] [Python]
Grover’s Algorithm offers a quadratic speedup in search problem, with respect to a classical exhaustive search. However, for structured problems, such as CSP (Constraint Satisfaction Problem), unbalanced search tree can lead to much faster classical algorithm. Algorithms, such as in the provided paper, suggested quantum speedup with respect to a structured classical search, using backtracking approach and quantum walks.
The project goal is to model a quantum backtracking algorithm based on one of the suggested approaches, using the Classiq platform, for a structured problem (CSP, SAT, etc.)
-  Smart Post Proccessing for Algorithms with QPE
prerequisites: [QAlgo] [Python]
Algorithms with Quantum Phase Estimation (QPE) are probably the most important use cases today for quantum computing. Those algorithms demand in the legacy approach circuit depth which is exponential in the wanted resolution of the result. Using Classiq’s platform we will demonstrate how can we get much better results by using much less data. The goal of this project is to learn the math theory behind this approach, demonstrate it on Finance use cases, and to explore together how it can be useful for other use cases such as Chemistry, Encryption or any idea by the students.
-  Designing native quantum arithmetic language
prerequisites: [QAlgo] [ProgLang]
Languages that encode quantum algorithms typically build up from a gate-level description. Classiq platform is unique in incorporating concepts at a much higher level of abstraction. Native arithmetic expressions are one important example. Arithmetic is used in search and optimization algorithms, among other applications.
In this project, we will explore the semantics and implementation of quantum arithmetic with the current Classiq platform. We will consider analogies from classical hardware description languages (HDLs), point out commonalities and differences between different uses of arithmetic, and try to formulate a more general abstract semantics.”
Variational Quantum Algorithm for the Fermi-Hubbard Model
prerequisites: [QM] [QAlgo]
The Fermi-Hubbard model is used constantly in condesed matter physics and material sciences. The ability to simulate it on a quantum computer and to extract relevant properties from it can offer a significant boost to many scientific domains.
In this project we will implement a variational quantum algorithm that efficiently computes ground-state properties of the Fermi-Hubbard model.
Links: Paper 1 & Paper 2
Symmetry-Efficient Ansatz for Variational Quantum Eigensolvers
prerequisites: [QM] [QAlgo]
Encoding smart ansatz that are tailored for a specific problem, is one a prominent direction for maximizing the performence of variational quantum algortrithms.
In this project, we will design an ansatz that preseerves the number of excitations of a quantum state, a common symmetry in physics and quantum chemistry. We will design the ansatz and ecorporate it with Classiq.
A student automatically fails if more than one progress is missing or she does not take part in writing paper or presentation.
We expect each student not only to dedicate enough time (quantity) but also to aim good outcomes (quality). Each student will collect points based on her activities during the QCourse.
- Completing the CLASSIQ Bootcamp is 12 points.
- Progress reports:
- A weekly regular work by a student is graded as 2 by the mentor. Less or more work is graded between 0 and 4.
- The bi-weekly progress is evaluated by two session guides. A standard outcome is graded as 3. Less or more progress is graded between 0 and 5 (from no progress to very good outcome). The scores by guides are averaged out.
- These two values are multiplied. So, a regular work with a standard progress is graded as 12 points, i.e., (2+2)*(3). Five similar progress is graded as 60 points.
- Writing paper: Equivalent to two progress report.
- A regular work by a student is graded as 4 by the mentor. Less or more work is graded between 0 and 8.
- A good paper is graded as 2 by a guide.
- These two values are multiplied. So, a regular work with a good paper is grades as 16 points.
- Final presentations: Equivalent to one progress report.
Thus regular work with standard outcomes will end with 108 points. A student collecting at least 70 points will pass Course551-1.
QCourse design: Abuzer Yakaryilmaz (QWorld & University of Latvia) & Eden Schirman (Classiq)
Coordinator: Abuzer Yakaryilmaz (QWorld & University of Latvia)
Guides: Eden Schirman (Classiq), Jibran Rashid (QWorld)
Guest guides: TBA
Mentors from Classiq: Israel Reichental, Lior Preminger, Eyal Cornfeld, Lior Gazit, Hanan Rosemarin, Nati Erez, Tal Michaeli, Gal Winer, Tamuz Danzig, Tomer Goldfriend, Or Samimi, Ron Cohen, and Matan Vax
Contact: qclass [at] qworld.net
Code of Conduct
Our course is dedicated to providing a harassment-free teaching and learning experience for everyone, regardless of gender, gender identity and expression, age, sexual orientation, disability, physical appearance, body size, race, ethnicity, religion (or lack thereof), or technology choices. We do not tolerate harassment of participants in any form. Sexual language and imagery are not appropriate for any event venue, including talks, workshops, parties, Twitter and other online media. Event participants violating these rules may be sanctioned or expelled from the course.
We respect the minors (children under age 18) and we must make every effort to protect their rights. All private relationships, private communications (including social media channels), or sexual contacts with minors are prohibited.
Except the filing the application form and similar formal procedures, the contact info of any attendee or participant cannot be requested by any person from organizer side (i.e., mentor, educator, speaker, organizer, sponsor, or volunteer). On the other hand, any person from organizer side may share his or her contact info with a participant who is not a minor, upon request by the participant.
A minor can access the emails of the main organizers on the event’s website. If a minor interested in working with a person from organizer side for scientific or pedagogical purpose, then he or she should read this document before contacting this person: https://qworld.net/code-of-ethics-and-conduct/#minors
If you are being harassed, notice that someone else is being harassed, or have any other concerns, please contact the course team immediately. For any concern regarding the course team, please contact the members of the Ethics Committee of QWorld. https://qworld.net/code-of-ethics-and-conduct/
Check the above link for more details.