Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Wednesday November 01 2017, @03:06PM   Printer-friendly
from the both-yes-and-no dept.

Submitted via IRC for SoyGuest31999

The hybrid study of quantum computers and artificial intelligence, or quantum machine learning, is still in its very early stages. Many of the machine learning algorithms are still theoretical and require large-scale quantum computers to be tested. Still, the marriage between the two has already proven fruitful.

[...] Quantum algorithms could solve problems within AI and vice versa. In addition, quantum theory could help develop more robust AI and help researchers formulate better algorithms. With this technology, we could see groundbreaking research in virtually all areas of study, from cancer research to geopolitics. We could use AI to help scientists solve incredibly nuanced and complex issues.

Source: https://venturebeat.com/2017/10/23/ai-and-quantum-algorithms-together-can-compute-a-better-world/


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 3, Insightful) by ese002 on Wednesday November 01 2017, @11:09PM

    by ese002 (5306) on Wednesday November 01 2017, @11:09PM (#590800)

    What sort of problems in AI can quantum computers solve?

    What sort of problems in physics can AI (as we currently understand it) solve?

    Look in TFA and you won't find answers to these questions. In fact, you won't even find the questions.

    About the closest is the assertion that AI needs a lot of compute power and quantum computers are really fast thus, good for AI. But the big problems in AI are not because of inadequate compute power and it is far from obvious that quantum computers can help AI problems that are related to compute performance. After all, quantum computing is not a universal cure for slow computation. It has been shown to work well for a few things, none of which (AFAIK) have anything to do with AI.

    Current AI amounts to increasingly sophisticated curve fitting. It can give excellent empirical results but provides no insight into correctness or underlying principles. It isn't obvious that it provides any utility toward improving our understanding of physics. That would seem to require a generalized learning system. We don't yet know how to build one of those even if it is allowed to run very slowly.

    Starting Score:    1  point
    Moderation   +1  
       Insightful=1, Total=1
    Extra 'Insightful' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   3