We invite you to a virtual workshop on CUDA-Q, NVIDIA’s platform for high-performance hybrid quantum-classical computing, featuring speakers and tutorials from NVIDIA, qBraid, and Infleqtion.
Useful quantum computing of the future will require the tight integration and interoperation of classical and quantum resources for quantum error correction, pre and post-processing, calibration, compilation, and applications. As such, CUDA-Q enables flexible and performant development of these workloads today, spanning CPUs, GPUs and QPUs, enabling quantum researchers, developers, and domain scientists to accelerate their quantum research at scale.
Part I
CUDA-Q Hands-on Tutorial:
In this virtual workshop, participants will learn how to develop hybrid applications using CUDA-Q, leveraging CUDA-Q’s powerful GPU-accelerated simulation libraries, in a hands-on environment. Participants will learn the following topics:
- CUDA-Q syntax using its user-friendly Pythonic interface
- How to develop quantum kernels and execute them on various backends, including single and multi-GPU
- How to develop noisy quantum circuits and custom noise models
- Hybrid quantum applications utilizing the GPU for classical and emulated quantum workloads
Part II
The following guest talks on CUDA-Q will be featured from qBraid and Infleqtion:
Accelerating Quantum R&D with CUDA-Q and NVIDIA GPUs through qBraid
Kenny Heitritter, Quantum Research Scientist, qBraid
In this presentation, we will provide an in-depth walkthrough of the qBraid platform, showcasing its value in quantum computing research and development. We will focus on demonstrating how to efficiently access CUDA-Q with seamless GPU integration through qBraid Lab. This session will include a live demo, allowing attendees to follow along using free qBraid accounts.
CUDA-Q enhanced Q-CHOP for Portfolio Optimization
Bharath Thotakura, Quantum Software Engineer, Infleqtion
Computing is an essential tool for the modern financial services industry. Indeed, massive profits are won and lost based on the speed and accuracy of the algorithms guiding financial decision making. Quantum computing has the potential to disrupt the financial services industry by integrating with state-of-the-art supercomputers and solving problems at scale that are otherwise impossibly complex. One such area is discrete portfolio optimization: a problem of selecting a portfolio of stocks that maximizes returns while minimizing volatility. Framed as a Quadratic Unconstrained Binary Optimization (QUBO), algorithms such as the Quantum Approximate Optimization Algorithm (QAOA) can be used to solve QUBOs on gate-based (universal) quantum computers, however, challenges remain with large scale problems. In this talk, we instead propose utilizing the Quantum Constrained Hamiltonian Optimization (Q-CHOP) algorithm, a novel method developed by Infleqtion that uses quantum properties to navigate the feasible solution space until it reaches the optimal constraint-satisfying solution. We will highlight the role of the Q-CHOP algorithm in tackling portfolio optimization and how our research was enabled by NVIDIA Quantum Cloud and CUDA-Q.
Access to GPUs
This is a hands-on workshop. To get the most out of the training, you will need to use NVIDIA GPUs. Therefore, space is limited and we will provide access to computing resources (GPUs) on a first-come, first-served basis. However, if you have your own resources or just want to listen in, you are also welcome. Register now!