[dead]
CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
141–150 of 166 posts
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#142They used a custom neural net with autoencoders, which contain convolutional layers. They trained it on previous experiment data. https://arxiv.org/html/2411.19506v1 Why is it so hard to elaborate what AI algorithm / technique they integrate? Would have made this article much better
Ah anomaly detection, that makes a lot more sense.
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#143I've got news for you, everybody with a modern cpu uses this, which use a perceptron for branch prediction.
Indeed, some examples: https://news.ycombinator.com/item?id=12340348 Neural network spotted deep inside Samsung's Galaxy S7 silicon brain (2016) https://ieeexplore.ieee.org/document/831066 Towards a high performance neural branch predictor (1999)
I'm borderline shocked that all of this extra overhead is somehow more efficient than something as simple as computing both branches or something.
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#144Earlier quoted context omitted.
It is 100% valid to label an algorithm that plays tic-tac-toe as "AI" Much of the early AI research was spent on developing various algorithms that could play board games. Didn't even need computers, one early AI was MENACE [1], a set of 304 matchboxes which could learn how to play noughts and crosses. [1] https://en.wikipedia.org/wiki/Matchbox_Educable_Noughts_and_...
I built the Matchbox for Hexapawn, detailed in National Geographic Kids! I didn't know what a Jujube was, but I got the idea.
That Hexapawn article was my first introduction to AI as a kid, though I never actually built it.
Found it in a "Reader's Digest Young Persons annual" which my dad got when he was a kid in the 60s. I still have that.
The original article from Scientific American: https://people.csail.mit.edu/brooks/idocs/GardnerHexapawn.pd...
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#145Not on the same extreme level, but I know that some coffee machines use a tiny CNN based model locally/embedded. There is a small super cheap camera integrated in the coffee machine, and the model does three things: (1) classifies the container type in order to select type of coffee, (2) image segmentation - to determine where the cup/hole is placed, (3) regression - to determine the volume and regulate how much coff…
Very cool, expensive machines?
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#146Earlier quoted context omitted.
I built the Matchbox for Hexapawn, detailed in National Geographic Kids! I didn't know what a Jujube was, but I got the idea.
That Hexapawn article was my first introduction to AI as a kid, though I never actually built it. Found it in a "Reader's Digest Young Persons annual" which my dad got when he was a kid in the 60s. I still have that. The original article from Scientific American: https://people.csail.mit.edu/brooks/idocs/GardnerHexapawn.pd...
You're probably talking about the same book I had then. I also remember it started off with a Mercury astronaut story, also had the story of Shackleton's Arctic voyage, and a two-page game board that was about trying to drive a car around China.
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#147I've got news for you, everybody with a modern cpu uses this, which use a perceptron for branch prediction.
Other news, is that HEP has used FPGAs for L0 triggers (amongst others) for decades. These always had a diverse selection criteria in their algorithms, event filters, suppression, weights etc. And just mentioning, that some custom radhard simple readout silicon from the likes of STM isn't any news either. And for historians: Delphi people (amongst others) had papers on Higgs selection using (A)NN from LEP data (overf…
It's a discussion forum. Saying people are all wrong with no proof comes off as arrogant but isn't helpful. If you have links to examples, you can simply say, "Here's some prior art or previous work in this area you all might like."
People would probably upvote that.
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#148the fact that 99% of LHC data is just gone forever is insane
Not really. Think of the experiment as a very, very high speed camera. They can't store every frame, so they try to capture just the "interesting" ones. They also store some random ones that can be used later as controls or in case they realize they've missed something. That's the whole job of these various layers of algorithms: recognizing interesting frames. Sometimes a new experiment basically just changes the def…
That we're building theories on what's left of mostly-trashed data has scientific implications. Most people hearing LHC proved something probably didn't think a preprocessor threw away most observations first. That layer of interpretation could cause errors.
I wonder how much independent review went into that step.
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#149Earlier quoted context omitted.
That Hexapawn article was my first introduction to AI as a kid, though I never actually built it. Found it in a "Reader's Digest Young Persons annual" which my dad got when he was a kid in the 60s. I still have that. The original article from Scientific American: https://people.csail.mit.edu/brooks/idocs/GardnerHexapawn.pd...
You're probably talking about the same book I had then. I also remember it started off with a Mercury astronaut story, also had the story of Shackleton's Arctic voyage, and a two-page game board that was about trying to drive a car around China.
Not the same. The story about John Glenn's "day in space" was about half way through, nothing about Shackleton and the two-page board game was "race to the moon"
The board game was paired with an abridged version of Wernher von Braun's "First Men to the Moon", which 7-year-old me assumed was an accurate depiction of the moon landings (literally in the same book as an actual Mercury mission, and the Challenger Deep dive).
They were probably pulling from the same pool of articles.
Re: CERN uses ultra-compact AI models on FPGAs for real-time LHC data filtering
#150Earlier quoted context omitted.
Some career do-nothing-but-make-noise in my organization hired a firm to 'Do AI' on some shitty data and the outcome was basically linear regression. It turns out that you can impressive executives with linear regression if you deliver it enthusiastically enough.
Not everyone knows everything so knowledge is the new oil. I do know about linear regression even had quite some of it at university. But I still wouldn’t be able to just implement it on some data without good couple days to weeks of figuring things out and which tools to use so I don’t implement it from scratch.
Implement it...from scratch? Its literally least squares regression. Its a few lines of code. What are you trying to say here?