By IQT News posted 03 Dec 2020

(Forbes) AI has already demonstrated its power across a range of use cases. But these powerful results aren’t necessarily delivered quickly. Machine learning (ML) models need to be trained, and in addition to demanding mountains of data, this takes time. Some models take months to train. But given the inherent advantages of quantum machines over classical machines, quantum computing can substantially reduce training times and improve the accuracy of ML models.
Recent research from IBM has borne out this quantum advantage for ML. Researchers demonstrated that, for certain classes of supervised ML models, a quantum-enabled ML algorithm performed not only faster but also more accurately than its classical computing counterpart. In practice, this means quantum-enabled ML will extract more valuable insights than even today’s most powerful classical computers and will be able to do so faster.
The generative capabilities of quantum devices have another potent application: filling in the gaps in data needed to train ML models. Data has traditionally been the limiting factor when it comes to training ML algorithms. That is, when trying to use ML to predict outcomes for rare or infrequent events, such as global pandemics, the absence or scarcity of a robust dataset poses significant challenges. Quantum-enabled generative AI models can create new “synthetic” datasets very similar to the training data you already have, greatly mitigating this limitation.

Subscribe to Our Email Newsletter

Stay up-to-date on all the latest news from the Quantum Technology industry and receive information and offers from third party vendors.

  • Forthcoming Events

    • IQT Fall | November 1-5, 2021
      Online & In-Person New York City
    • IQT, The Hague, The Netherlands | February 21-23, 2022
    • IQT Spring | May 10-13, 2022
      Online & In-Person San Diego
    • IQT Asia-Pacific Singapore 2022 | Dates Forthcoming
    • For additional information: info@3drholdings.com
0