A New Kind of Computer (April 2025)
lightmatter.co
A New Kind of Computer (April 2025)
1–10 of 27 posts
Re: A New Kind of Computer (April 2025)
#2Re: A New Kind of Computer (April 2025)
#3Hmm... what? So it is not accurate?
Re: A New Kind of Computer (April 2025)
#4I stopped reading after "Soon, you will not be able to afford your computer. Consumer GPUs are already prohibitively expensive."
Re: A New Kind of Computer (April 2025)
#5Maybe not the specific photonic system that they are describing. Which I'm sure has some significant improvements over what existed then. But the idea of using analog approximations of existing neural net AI models, to allow us to run AI models far more cheaply, with far less energy.
Whether or not this system is the one that wins out, I'm very sure that AI run on an analog system will have a very important role to play in the future. It will allow technologies like guiding autonomous robots with AI models running on hardware inside of the robot.
Re: A New Kind of Computer (April 2025)
#6> Critically, this processor achieves accuracies approaching those of conventional 32-bit floating-point digital systems “out-of-the-box,” without relying on advanced methods such as fine-tuning or quantization-aware training. Hmm... what? So it is not accurate?
However a single analog math operation requires the same energy as a single bit flip in a digital computer. And it takes a lot of bit flips to do a single floating point operation. So a digital calculation can be approximated with far less energy and hardware. And neural nets don't need digital precision to produce useful results.
Re: A New Kind of Computer (April 2025)
#7I stopped reading after "Soon, you will not be able to afford your computer. Consumer GPUs are already prohibitively expensive."
Re: A New Kind of Computer (April 2025)
#8> Critically, this processor achieves accuracies approaching those of conventional 32-bit floating-point digital systems “out-of-the-box,” without relying on advanced methods such as fine-tuning or quantization-aware training. Hmm... what? So it is not accurate?
It's an analog system. Which means that accuracy is naturally limited. However a single analog math operation requires the same energy as a single bit flip in a digital computer. And it takes a lot of bit flips to do a single floating point operation. So a digital calculation can be approximated with far less energy and hardware. And neural nets don't need digital precision to produce useful results.
The point - as shown by the original implementation...