From NVIDIA: 2024-10-08 11:00:00

A new study by NVIDIA, Moderna, and Yale suggests that quantum machine learning can improve drug discovery by predicting molecular properties more accurately. This advancement could lead to more efficient pharmaceutical therapies. The study emphasizes the use of GPU-accelerated simulation of quantum algorithms in exploring these methods for enhancing AI techniques in drug discovery.

Researchers are investigating how quantum neural networks can impact real-world use cases like drug discovery. Large-scale simulations of future noiseless quantum processing units are needed to study the effects of quantum computing on drug discovery and other complex tasks. This research highlights the increasing importance of GPU-accelerated supercomputing in tackling challenges as quantum computing scales up.

NVIDIA’s CUDA-Q quantum development platform enables multi-GPU accelerated simulations of quantum machine learning workloads. The platform allows for the simultaneous simulation of multiple quantum processing units, making it ideal for studying realistic large-scale devices and conducting quantum machine learning tasks that batch training data.

The study covers various quantum machine learning techniques, including hybrid quantum convolutional neural networks, which require the ability to write programs combining classical and quantum resources. NVIDIA’s involvement in developing practical quantum computers is evident in the increased reliance on GPU supercomputing demonstrated in this research. NVIDIA will further showcase its role in quantum computing at the upcoming SC24 conference in Atlanta, Nov. 17-22.



Read more at NVIDIA: Accelerated Computing Key to Yale’s Quantum Research