Topic outline

  • General

    Scope: 5 cr

    Languages ​​of performance: English

    Evaluation scale: General scale, 0-5

    Course level: Advanced Studies

    Instructors: Vikas Garg and Alexandru Paler. 

    Background:
    Quantum methods can provide significant speedup over classical algorithms for problems such as searching for an item in a large list. Quantum Machine Learning (QML) is an emerging field that strives to combine the strengths of Quantum Computing and Machine Learning. One line of work in QML has focused on devising quantum analogues, of standard machine learning algorithms, that can be executed on a quantum computer (as and when quantum hardware become ready to work with large datasets).  Another line of work attempts to draw insights from quantum methods to design better algorithms for machine learning tasks that can be executed on the standard (non-quantum) computing machines. Finally, a hybrid philosophy advocates algorithms  that require executing only a part of the model on quantum hardware, delegating the rest to the classical hardware.

    Objectives: 

    Understanding the general principles of quantum computing and quantum machine learning. The students are expected to be able to design quantum learning methods and circuits for some classical machine learning algorithms, and to conduct research in quantum machine learning and/or apply quantum methods on real datasets.

    Structure:  Students would work in small teams on a project, and shall be evaluated on their performance throughout the course (meetings, updates, implementation, documentation, and a final presentation).  The students would be provided a list of projects to work on, but are welcome to suggest their own projects as well. A suggestive list of areas for the project is here:

    -       Adiabatic Computing
    -       Hybrid Methods
    -       Quantum Circuit Optimization 
    -       Generative Models (GANs, VAEs, etc.)
    -       Applications of Quantum Learning Methods

     Prerequisites:  Mathematical maturity.  Prior background in quantum computing would be helpful, but is not mandatory. The course would be ideal for those having interest in theoretical computer science, applied math, applications of quantum computing, or machine learning.

    Registration: Since this is an advanced course-based project, where students work in small teams, the number of registrations will be capped. The interested students should express their interest by sending an email containing their CV as an attachment to (vikas.garg@aalto.fi), and will be notified shortly afterwards about their enrolment in the course.