Webinars

September 4, 2019 at 2PM EDT
Webinar on "A gentle introduction to quantum computation in the statistical sciences" by Mark Fingerhuth

Abstract
Supercomputers based on classical mechanics have contributed to major breakthroughs in all aspects of modern life but big data, large-scale simulations and hard optimization problems frequently push today's hardware to its limits. Quantum computers are a new supercomputing paradigm that is based on the fundamentals of quantum mechanics and which has the potential to enable dramatic speedups over classical hardware. The theoretical idea of building a quantum computer was first published in the 80s and in the last decade the field has seen tremendous breakthroughs. Three years ago, the first quantum computer became publicly accessible and this year we are expecting the first demonstration of what is known as quantum advantage; a problem that can be solved faster and/or more efficiently on quantum hardware rather than on a classical supercomputer.

The field of quantum computation provides new algorithms and approaches to problems such as combinatorial optimization, sampling and machine learning, thereby impacting a wide variety of fields i.e. finance, material science, supply chain, life sciences and manufacturing. Quantum algorithms that are meant to run on quantum hardware exclusively have been shown to provide exponential speedups whilst others also make use of classical hardware through quantum/classical hybrid architectures. In this webinar, you will learn about the basic building blocks of quantum computers and the different types of algorithms and their applications.

Next, we will focus on a particular quantum algorithm with an application in the life sciences. Big pharma frequently employs computational techniques such as shallow classifiers, deep neural networks and enhanced sampling algorithms for drug discovery in order to perform rational design and optimization of therapeutic molecules. When studying the dynamics of drug targets as well as protein therapeutics, one needs to be able to predict the three dimensional structure of an amino acid sequence - widely known as the protein folding problem. Most approaches to protein folding are either based on protein fragment libraries, deep neural networks or Monte-Carlo techniques. The main challenge is that protein folding is an NP-hard problem which implies computationally intractability on classical supercomputers. In the second half of the webinar, I will walk you step-by-step through a quantum/classical hybrid algorithm for protein folding, outline the workflow of programming a quantum computer and provide you with a framework of how to evaluate life science problems in the context of quantum computing.


About the speaker:
Mark Fingerhuth is a theoretical physicist by training and entrepreneur by choice. In late 2017, he co-founded the Canadian biotech startup ProteinQure which revolutionizes the computational design of protein therapeutics. ProteinQure is one of the few companies worldwide that pioneers the use of quantum computing technology in the life sciences. Academically, he published one of the first proof-of-principle implementation of a quantum machine learning algorithm on superconducting quantum hardware and has co-authored several papers on quantum software, quantum memory as well as quantum-classical hybrid algorithms. Together with Peter Wittek and Tomas Babej, he recently co-founded the Quantum Open Source Foundation which supports and promotes the development of free and open quantum software through conferences and hackathons.

For further details about the webinar, please contact Sergei.Leonov@cslbehring.com