6 thoughts on “Deep belief networks

  1. Yes, this is a great example of physics and computer science intersecting. We tend to think of the quantum computing paradigm being the archetype but in this case it’s stat mech that provides the interface. This example is much closer to AQC and much learnings can be had for folks interested in physics / comp sci / AQC by investigating these types of approaches. I think this area could be a gold mine for quantum computation.

  2. looks like he needs some help with images, I guess that where dwave steps in.😉

    new levels of hardware needs/creates new levels of software. I think once your hardware you’ll be surprised the different way people come up with using it

  3. Better captcha cracking through science! I love it. With your AQCs and my spam network, we can RULE THE WORLD! Or at least sell it Viagra and breast implants. The future is bright indeed, my friend.🙂

  4. My favourite quote is “If you choose a bigger number, then PCA [(Principal Component Analysis)] will do better, so your comparison with PCA won’t be as good.”

    Unfortunately for him, PCA is *extremely* simple and used for just about everything, so unless that time scaling improvement he mentioned is relative to PCA, he’s got an uphill battle for demonstrating robustness of his approach. I’m not a fan of the pick a random number and hope that it works well for that number technique, so hopefully someone will figure out a good way of finding the magic number for different tasks.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s