A not too long ago launched joint analysis paper by NVIDIA, Moderna and Yale evaluations how methods from quantum machine studying (QML) might improve drug discovery strategies by higher predicting molecular properties.
Finally, this might result in the extra environment friendly era of latest pharmaceutical therapies.
The evaluate additionally emphasizes that the important thing instrument for exploring these strategies is GPU-accelerated simulation of quantum algorithms.
The research focuses on how future quantum neural networks can use quantum computing to reinforce current AI methods.
Utilized to the pharmaceutical business, these advances provide researchers the power to streamline complicated duties in drug discovery.
Researching how such quantum neural networks affect real-world use circumstances like drug discovery requires intensive, large-scale simulations of future noiseless quantum processing items (QPUs).
This is only one instance of how, as quantum computing scales up, an rising variety of challenges are solely approachable with GPU-accelerated supercomputing.
The evaluate article explores how NVIDIA’s CUDA-Q quantum growth platform gives a singular instrument for operating such multi-GPU accelerated simulations of QML workloads.
The research additionally highlights CUDA-Q’s means to simulate a number of QPUs in parallel. It is a key means for learning practical large-scale gadgets, which, on this explicit research, additionally allowed for the exploration of quantum machine studying duties that batch coaching information.
Lots of the QML methods coated by the evaluate — resembling hybrid quantum convolution neural networks — additionally require CUDA-Q’s means to jot down applications interweaving classical and quantum assets.
The elevated reliance on GPU supercomputing demonstrated on this work is the newest instance of NVIDIA’s rising involvement in creating helpful quantum computer systems.
NVIDIA plans to additional spotlight its function in the way forward for quantum computing on the SC24 convention, Nov. 17-22 in Atlanta.