Article Image
IPFS News Link • Science, Medicine and Technology

QC Ware Quantum Loader Breakthrough for up 100X Faster Training of Quantum Machine Learning

•, by Brian Wang

They have achieved a speedup of about 100 times.

Nextbigfuture interviewed Yianni Gamvros, Head of Product and Business Development at QC Ware. Quantum Computers are still much slower than classical computers. However, the new loading system, algorithm improvements and hardware improvements are rapidly enabling quantum systems to catch up to classical computers.

Classical AI systems have improved 44 times over the last eight years using algorithmic improvement. ImageNet classification has been decreasing by a factor of 2 every 16 months. Compared to 2012, it now takes 44 times less compute to train a neural network to the level of AlexNet. Moore's Law would yield an 11x cost improvement over this period with hardware speedup. there are custom AI chips that have about 100X speedup.

QC Ware's algorithms researchers can efficiently load classical data onto quantum hardware and how distance estimations can be performed quantumly. These new capabilities enabled by Data Loaders are now available in the latest release of QC Ware's Forge™ cloud services platform, an integrated environment to build, edit, and implement quantum algorithms on quantum hardware and simulators.