Mastodawn

TORcicada Mar 28

CERN uses tiny AI models burned into silicon for real-time LHC data filtering

https://theopenreader.org/Journalism:CERN_Uses_Tiny_AI_Models_Burned_into_Silicon_for_Real-Time_LHC_Data_Filtering

CERN Uses Tiny AI Models Burned into Silicon for Real-Time LHC Data Filtering

CERN’s deployment of the 'AXOL1TL' (Anomaly eXploration with On-chip L1 Trigger) algorithm represents a critical shift toward edge-computing at the sub-atomic scale. By utilizing the 'hls4ml' transpiler to map deep learning architectures directly onto radiation-hard FPGAs and custom ASICs, the Level-1 trigger achieves nanosecond-scale inference (under 50ns) to filter a 40MHz collision stream. The architecture bypasses traditional Von Neumann bottlenecks through extreme quantization (2-6 bit weight precision) and the extensive use of precomputed lookup tables (LUTs), effectively 'burning' the model logic into the physical substrate. This hardware-first approach enables the real-time identification of 1-in-a-trillion 'Rare Signal' events while discarding 99.98% of the 40,000 exabytes of annual background noise.

TheOpenReader

They 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

Real-time Anomaly Detection at the L1 Trigger of CMS Experiment

dcanelhas Mar 28

I'm half expecting to see "AI model" appearing as stand-in for "linear regression" at this point in the cycle.

I'm sure I've seen basic hill climbing (and other optimisation algorithms) described as AI, and then used evidence of AI solving real-world science/engineering problems.

LiamPowell Mar 28

Historically this was very much in the field of AI, which is such a massive field that saying something uses AI is about as useful as saying it uses mathematics. Since the term was first coined it's been constantly misused to refer to much more specific things.

From around when the term was first coined: "artificial intelligence research is concerned with constructing machines (usually programs for general-purpose computers) which exhibit behavior such that, if it were observed in human activity, we would deign to label the behavior 'intelligent.'" [1]

[1]: https://doi.org/10.1109/TIT.1963.1057864

That definition moves the goalposts almost by definition, people only stopped thinking that chess demonstrated intelligence when computers started doing it.

The term artificial intelligence has always been just a buzzword designed to sell whatever it needed to. IMHO, it has no meaningful value outside of a good marketing term. John McCarthy is usually the person who is given credit for coming up with the name and he has admitted in interviews that it was just to get eyeballs for funding.

ninjagoo Mar 28

> I'm half expecting to see "AI model" appearing as stand-in for "linear regression" at this point in the cycle.

Already the case with consulting companies, have seen it myself

I'm half expecting to see "AI model" appearing as stand-in for "if > 0" at this point in the cycle.

Foobar8568 Mar 28

This is why I am programming now in Ocaml, files themselves are AI ( ml ).

I am sure you did not forget that pattern matching.

Having work with people who do that, I can guarantee that’s not the case.
See https://ssummers.web.cern.ch/conifer/ and HSL4ML, these run BDT and CNN

API Reference — conifer 1.8 documentation

That works well to get around patents btw :)

Because if it’s not an LLM it’s not good for the current hype cycle. Calling everything AI makes the line go up.

danielbln Mar 28

LLMs also make the cynicism go up among the HN crowd.

etrautmann Mar 28

It seems like most of the implementation is FPGA, which I wouldn’t call “physically burned into silicon.” That’s quite a stretch of language