News for hackers. Stuff that matters.
Re:Making floating point math highly efficient for AI hardware by Geee on Friday November 09, 2018 @10:49PM (#18419821)
A bit off-topic, but I remember some studies about 'under-powered' ASICs, ie. running with 'lower-than-required' voltage and just letting the chip fail sometimes. I guess the outcome was that you can run with 0.1x power and get 0.9x of correctness. Usually chips are designed so that they never fail and that requires using substantially more energy than is needed in the average case. If the application is probabilistic or noisy in general, additional 'computation noise' could be allowed for better energy efficiency.
Re:Making floating point math highly efficient for AI hardware by david-gpu on Saturday November 10, 2018 @12:24PM (#18422288)
That sounds awful for verification, debugging, reproducibility and safety-critical systems. Imagine this in a self-driving car. Scary.
Re:Making floating point math highly efficient for AI hardware by TomMarius on Monday November 12, 2018 @07:32AM (#18432209)
Well, you could simply not use these in a self-driving vehicle.