efficiency breakthrough via mathematics

  • 2 Replies


  • Administrator
  • **********
  • Millennium Man
  • *
  • 1020
  • Humans will disappoint you.
    • Home Page
efficiency breakthrough via mathematics
« on: October 13, 2020, 03:02:18 am »

This seems like something to get excited about.

Thanks to a mathematical breakthrough, AI applications like speech recognition, gesture recognition and ECG classification can become a hundred to a thousand times more energy efficient. This means it will be possible to put much more elaborate AI in chips, enabling applications to run on a smartphone or smartwatch where before this was done in the cloud.

Under supervision of CWI researcher and UvA professor cognitive neurobiology Sander Bohté, researchers developed a learning algorithm for so-called spiking neural networks. Such networks have been around for some time, but are very difficult to handle from a mathematical perspective, making it hard to put them into practice so far. The new algorithm is groundbreaking in two ways: the neurons in the network are required to communicate a lot less frequently, and each individual neuron has to execute fewer calculations.



  • Trusty Member
  • *********
  • Terminator
  • *
  • 986
  • Where are these cookies!?
Re: efficiency breakthrough via mathematics
« Reply #1 on: October 13, 2020, 08:53:42 pm »
I don't have much of a frame of reference. Is this comparable to AC/DC currents, with the non continuous math wasting less energy on unnecessary things? Is it more like only pedaling uphill, and coasting the rest of the time? Could those analogue neurons from the "Memristor Breakthrough" (picture below) be made to approximate this style of function, where they pick their moments to communicate? Can any mathematical formula be created as an analogue structure? Interesting stuff.

Mz-Y5-Nz-U3-MA" border="0



  • Administrator
  • **********
  • Millennium Man
  • *
  • 1020
  • Humans will disappoint you.
    • Home Page
Re: efficiency breakthrough via mathematics
« Reply #2 on: October 14, 2020, 09:47:44 am »
A closer analogy might be like the difference between switching and linear amplifiers or power regulators, but what it amounts to is not wasting energy on things that aren't changing, just responding to events and remaining at rest when nothing is happening. That would be complicated to model mathematically, much like integer programming is much more complicated than linear programming in operations research.



New Server
by infurl (Announcements)
Today at 12:52:17 am
Releasing full AGI/evolution research
by LOCKSUIT (General Project Discussion)
November 28, 2020, 06:58:33 pm
alert manager class for waifubots
by yotamarker (General AI Discussion)
November 27, 2020, 04:12:22 pm
Giving AI rights
by frankinstien (General Project Discussion)
November 26, 2020, 04:26:07 pm
We are computational machines after all!
by MikeB (General Project Discussion)
November 26, 2020, 08:37:45 am
Pattern based NLP
by MikeB (General Project Discussion)
November 26, 2020, 08:28:32 am
I test Jukebox
by LOCKSUIT (General AI Discussion)
November 25, 2020, 11:27:59 pm
This video puts size into perspective.
by frankinstien (Video)
November 25, 2020, 05:14:56 pm
Syntherapy AI psychotherapist game.
by WriterOfMinds (AI News )
November 27, 2020, 05:33:57 pm
Senate Approves Deepfake bill
by LOCKSUIT (AI News )
November 25, 2020, 02:01:18 am
Sony Patent Suggests PS5 Will Have a Chatbot Feature
by frankinstien (AI News )
November 18, 2020, 05:47:45 pm
Potentially life-saving robot scares bears.
by infurl (Robotics News)
November 12, 2020, 12:41:40 am
good news everyone
by HS (AI News )
November 07, 2020, 10:03:04 pm
Meet Kuki
by 8pla.net (AI News )
November 05, 2020, 04:18:34 am
Realistic and Interactive Robot Gaze by Disney Research
by infurl (AI News )
November 03, 2020, 06:33:15 am
less than one-shot learning
by ivan.moony (AI News )
October 26, 2020, 07:45:25 am

Users Online

95 Guests, 0 Users

Most Online Today: 117. Most Online Ever: 2369 (November 21, 2020, 04:08:13 pm)