I must be old because first thing I thought reading AlphaChip was why is deepmind talking about chips in DEC Alpha :-) https://en.wikipedia.org/wiki/DEC_Alpha .
haha, same!
How AlphaChip transformed computer chip design
21–30 of 215 posts
Re: How AlphaChip transformed computer chip design
#22How far are we from memory-based computing going from research into competitive products? I get the impression that we are already well passed the point where it makes sense to invest very aggressively to scale up experiments with things like memristors. Because they are talking about how many new nuclear reactors they are going to need just for the AI datacenters.
Re: How AlphaChip transformed computer chip design
#23Earlier quoted context omitted.
Floorplanning algorithms and solvers already exist https://limsk.ece.gatech.edu/course/ece6133/slides/floorplan...
The original paper from DeepMind evaluates what they are now calling AlphaChip versus existing optimizers, including simulated annealing. They conclude that AlphaChip outperforms them with much less compute and real time. https://www.cl.cam.ac.uk/~ey204/teaching/ACS/R244_2021_2022/...
Randomized algorithms strike again!
Re: How AlphaChip transformed computer chip design
#24A marvellous achievement from DeepMind as usual, I am quite surprised that Google acquired them for a significant discount of $400M, when I would have expected it to be in the range of $20BN, but then again Deepmind wasn’t making any money back then.
Re: How AlphaChip transformed computer chip design
#25Looks like this is only about placement. I wonder if it can be applied to routing?
Exactly what I was thinking.
Also: when is this coming to KiCad? :)
PS: It would also be nice to apply a similar algorithm to graph drawing (e.g. trying to optimize for human readability instead of electrical performance).
Re: How AlphaChip transformed computer chip design
#26I understand the achievement, but can't square it with my belief that uniform systolic arrays will prove to be the best general purpose compute engine for neural networks. Those are almost trivial to route, by nature.
Isn't this already the case for large portions of GPUs? Like, many of the blocks would be systolic arrays?
I think the next step is arrays of memory-based compute.
Re: How AlphaChip transformed computer chip design
#27Can this be abstracted and generalized into a more generally applicable optimization method?
Re: How AlphaChip transformed computer chip design
#28Every generation of chips is used to design next generation. That seems to be the root of exponential growth in Moore's law.
Re: How AlphaChip transformed computer chip design
#29Earlier quoted context omitted.
The original paper from DeepMind evaluates what they are now calling AlphaChip versus existing optimizers, including simulated annealing. They conclude that AlphaChip outperforms them with much less compute and real time. https://www.cl.cam.ac.uk/~ey204/teaching/ACS/R244_2021_2022/...
Randomized algorithms strike again!
This is moreso amortized optimization/reinforcement learning, not randomized algorithms.
Re: How AlphaChip transformed computer chip design
#30I hope I'll still be alive when they'll announce AlephZero.