Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How AlphaChip transformed computer chip design

deepmind.google

Post reply on HN
Print view View original

191–200 of 215 posts

Re: How AlphaChip transformed computer chip design

#191

Post by negativeonehalf »

dogleg77 wrote: post #186
negativeonehalf wrote: post #71

Earlier quoted context omitted.

FD: I have been following this whole thing for a while, and know personally a number of the people involved. The AlphaChip authors address criticism in their addendum, and in a prior statement from the co-lead authors: https://www.nature.com/articles/s41586-024-08032-5 , https://www.annagoldie.com/home/statement - The 2023 ISPD paper didn't pre-train at all. This means no learning from experience, for a learning-base…

The problem with the Google Nature paper is that its results were not reproduced outside Google. You can attack attempts to reproduce but that only reinforces the point: those claimed results cannot be trusted. Other commenters already addressed the pre-training issue. Please kindly include a link to Kahng's 2023 discussion addressing your complaints. Otherwise, you are unfairly supporting those people you know. Kahn…

For a more thorough discussion on pre-training, see this ISPD 2022 paper by the AlphaChip people: https://dl.acm.org/doi/pdf/10.1145/3505170.3511478

As for external usage of the method - MediaTek is one of the largest chip design companies in the world, and they built on AlphaChip. There's a quote from a MediaTek SVP at the bottom of the GDM blog post:

"AlphaChip's groundbreaking AI approach revolutionizes a key phase of chip design. At MediaTek, we've been pioneering chip design's floorplanning and macro placement by extending this technique in combination with the industry's best practices. This paradigm shift not only enhances design efficiency, but also sets new benchmarks for effectiveness, propelling the industry towards future breakthroughs."

Re: How AlphaChip transformed computer chip design

#192

Post by negativeonehalf »

AshamedCaptain wrote: post #127

What is Google doing here? At best, the quality of their "computer chip design" work can be described as "controversial" https://spectrum.ieee.org/chip-design-controversy . What is there to gain by just making a PR now without doing anything new?

In the blog post, they announce MediaTek's widespread usage, the deployment in multiple generations of TPU with increasing performance each generation, Axion, etc.

Chips designed with the help of AlphaChip are in datacenters and Samsung phones, right now. That's pretty neat!

Re: How AlphaChip transformed computer chip design

#193

Post by wegfawefgawefg »

317070 wrote: post #157
wegfawefgawefg wrote: post #124

Earlier quoted context omitted.

In reinforcement learning pre-training reduces peak performance. We can argue about this, but it is not a sufficiently strong point to stop reading from alone.

Do you have a citation for this? I did my Phd on this topic 8 years ago, and I didn't completely follow the field after. I'm curious to learn more.

Not off the top of my head, but back when self play was first being figured out the competing strategy was behavioural cloning, and there was some flirting with bootstrapping self play with initial behavioural cloning. It would always bias the policy and reduce exploration. You end up with a worse final policy. Best to train from scratch. All the top rl papers did no behavioural pretraining and beat out the ones that did by many orders of magnitude on scores.

We are going to relearn this lesson with ambulation and grasping as all the large companies are trying to make useful robots from human shadowing to reduce the gigantic sample size requirements burden with self play. Likely after the initial years computers will just get a couple more doublings in compute per watt and we will see the full self training models take over those domains as the old human data biased models get thrown out.

Re: How AlphaChip transformed computer chip design

#194

Post by sroussey »

ilaksh wrote: post #105
foota wrote: post #104

Earlier quoted context omitted.

I don't think it's ignorant to not invest in something that has a decade long path towards even having a market, much less a large market.

I have seen at least one experiment running a language model or other neural network on (small scale) memory-based computing substrates. That suggests less than 1-2 years to apply them immediately to existing tasks once they are scaled up in terms of compute capacity.

Many more years than that. And it must be general enough. Otherwise you optimize for A in hardware and 5yr later when producing chips, A is no longer revenant and everyone moved to B.

Re: How AlphaChip transformed computer chip design

#195

Post by sijnapq »

negativeonehalf wrote: post #191
dogleg77 wrote: post #186

Earlier quoted context omitted.

The problem with the Google Nature paper is that its results were not reproduced outside Google. You can attack attempts to reproduce but that only reinforces the point: those claimed results cannot be trusted. Other commenters already addressed the pre-training issue. Please kindly include a link to Kahng's 2023 discussion addressing your complaints. Otherwise, you are unfairly supporting those people you know. Kahn…

For a more thorough discussion on pre-training, see this ISPD 2022 paper by the AlphaChip people: https://dl.acm.org/doi/pdf/10.1145/3505170.3511478 As for external usage of the method - MediaTek is one of the largest chip design companies in the world, and they built on AlphaChip. There's a quote from a MediaTek SVP at the bottom of the GDM blog post: "AlphaChip's groundbreaking AI approach revolutionizes a key phas…

[dead]

Re: How AlphaChip transformed computer chip design

#196

Post by sijnapq »

negativeonehalf wrote: post #192
AshamedCaptain wrote: post #127

What is Google doing here? At best, the quality of their "computer chip design" work can be described as "controversial" https://spectrum.ieee.org/chip-design-controversy . What is there to gain by just making a PR now without doing anything new?

In the blog post, they announce MediaTek's widespread usage, the deployment in multiple generations of TPU with increasing performance each generation, Axion, etc. Chips designed with the help of AlphaChip are in datacenters and Samsung phones, right now. That's pretty neat!

[dead]

Re: How AlphaChip transformed computer chip design

#197

Post by sijnapq »

negativeonehalf wrote: post #190
gabegobblegoldi wrote: post #109

Earlier quoted context omitted.

Vindicated indeed. The senior researcher and others on the project were bullied for raising concerns of fraud by the two researchers [1]. They filed a lawsuit against Google that has a lot of detailed allegations of fraud [2]. [1] https://www.theregister.com/AMP/2023/03/27/google_ai_chip_pa... [2] https://regmedia.co.uk/2023/03/26/satrajit_vs_google.pdf

You are now using multiple new accounts based on the name of one of the authors (Anna Goldie) and her husband (Gabriel). First this one ('gabegobblegoldi'), and then 'anna-gabriella'. I think it is time for you to take a deep breath and think about what you are doing and why. You seem to be obsessed with the idea that this work is overrated. MediaTek and Google don't think so, and use it in production for their chips…

[dead]

Re: How AlphaChip transformed computer chip design

#198

Post by anna-gabriella »

negativeonehalf wrote: post #190
gabegobblegoldi wrote: post #109

Earlier quoted context omitted.

Vindicated indeed. The senior researcher and others on the project were bullied for raising concerns of fraud by the two researchers [1]. They filed a lawsuit against Google that has a lot of detailed allegations of fraud [2]. [1] https://www.theregister.com/AMP/2023/03/27/google_ai_chip_pa... [2] https://regmedia.co.uk/2023/03/26/satrajit_vs_google.pdf

You are now using multiple new accounts based on the name of one of the authors (Anna Goldie) and her husband (Gabriel). First this one ('gabegobblegoldi'), and then 'anna-gabriella'. I think it is time for you to take a deep breath and think about what you are doing and why. You seem to be obsessed with the idea that this work is overrated. MediaTek and Google don't think so, and use it in production for their chips…

[flagged]

Re: How AlphaChip transformed computer chip design

#199

Post by sijnapq »

negativeonehalf wrote: post #190
gabegobblegoldi wrote: post #109

Earlier quoted context omitted.

Vindicated indeed. The senior researcher and others on the project were bullied for raising concerns of fraud by the two researchers [1]. They filed a lawsuit against Google that has a lot of detailed allegations of fraud [2]. [1] https://www.theregister.com/AMP/2023/03/27/google_ai_chip_pa... [2] https://regmedia.co.uk/2023/03/26/satrajit_vs_google.pdf

You are now using multiple new accounts based on the name of one of the authors (Anna Goldie) and her husband (Gabriel). First this one ('gabegobblegoldi'), and then 'anna-gabriella'. I think it is time for you to take a deep breath and think about what you are doing and why. You seem to be obsessed with the idea that this work is overrated. MediaTek and Google don't think so, and use it in production for their chips…

[dead]

Re: How AlphaChip transformed computer chip design

#200

Post by gabegobblegoldi »

negativeonehalf wrote: post #102
Drunk_Engineer wrote: post #96

Earlier quoted context omitted.

From what I saw in the rebuttal papers, the Google cost-function is wirelength based. You can still get good TNS from that if your timing is very simplistic -- or if you choose your benchmark carefully.

They optimize using a fast heuristic based on wirelength, congestion, and density, but they evaluate with full P&R. It is definitely interesting that they get good timing without explicitly including it in their reward function!

Interesting == Suspicious? I think this is a big red flag to those in the know.
Post reply on HN