Missing momentum is effectively the only way that the main detectors at the LHC (like CMS) can measure the indirect signs of neutrinos (or undiscovered particles). This #CMSPaper present how we use #machinelearning to make our missing momentum measurement much, much better: arxiv.org/abs/2509.12012
📣 We’re recruiting Volunteer TA Selection Committee Members
This role is ideal for academics, educators, and researchers who understand Python and who care deeply about teaching quality, mentorship, and equitable access to graduate-level computational education. If you are a previous NMA/CMA student or TA, we’d love to hear from you!
Applications close 15 Feb
➡️ Learn more and apply here: https://neuromatch.io/volunteer/
#ComputationalScience #Neuroscience #MachineLearning #DeepLearning #NeuroAI
4 media
This #CMSPaper investigates different #AI #machinelearning methods that aim to find jets that are inconsistent with the standard model. It shows that a new method called #Wasserstein normalized autoencodes works much better than other neural networks arxiv.org/abs/2510.02168
Alex, the Hearth Fire
boosted
As much as I loathe LLM "AI" built from hoards of stolen data, machine learning "AI" has become terrifically useful.
This past week I had 10 audio recorders set out in the forest and nearby grassland, all recording non-stop from Monday afternoon to Friday morning. That was on our recent university field ecology field trip.
Today I downloaded all the files to a hard drive (156 GB of data) and then I set my little M1 Macbook Air to work, using the offline desktop BirdNet app to identify all of the birds in the recordings.
It took most of the day, and now I have a 42,284 row spreadsheet of birds detected.
It really feels like magic.
Here's a quick sorted lists of all the bird detections with species IDs with a confidence score >0.9.
Together with the students in the course, we'll later compare how birds have changed since we started doing this in 2020, and how the birds in the grassland differ from the forest.
As much as I loathe LLM "AI" built from hoards of stolen data, machine learning "AI" has become terrifically useful.
This past week I had 10 audio recorders set out in the forest and nearby grassland, all recording non-stop from Monday afternoon to Friday morning. That was on our recent university field ecology field trip.
Today I downloaded all the files to a hard drive (156 GB of data) and then I set my little M1 Macbook Air to work, using the offline desktop BirdNet app to identify all of the birds in the recordings.
It took most of the day, and now I have a 42,284 row spreadsheet of birds detected.
It really feels like magic.
Here's a quick sorted lists of all the bird detections with species IDs with a confidence score >0.9.
Together with the students in the course, we'll later compare how birds have changed since we started doing this in 2020, and how the birds in the grassland differ from the forest.
#MachineLearning anlysis of 20 years of in situ measurements in the #Earth's #magnetosphere brings new insights on #MagneticReconnection processes:
📄 https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2025GL119118
#SpacePhysics #Plasma #PlasmaPhysics #SolarWind #Reconnection @irap https://social.numerique.gouv.fr/@irap/115966776951772619
Jan :rust: :ferris:
boosted
In the Reverse Engineering world, we have a rule: You don't own it until you can take it apart.
The same applies to Artificial Intelligence.
We are currently drowning in API wrappers. Everyone is "building AI apps," but very few people are looking at the wiring underneath. To truly understand modern LLMs, I decided to stop using libraries. I went back to the drawing board to build a custom architecture from scratch.
Meet SARAN (Shallow Auto-Regressive Attention Network).
It’s not designed to beat GPT-4. It’s designed to be transparent. 🔹 I built a strict 15-stage computational graph. 🔹 I manually implemented backpropagation to trace the gradients. 🔹 I scaled it to a 354M parameter model to watch how it learns.
I’ve documented the entire build log—including the architecture decisions and the "why" behind the math—in my new engineering newsletter, Bits & Neurons.
If you want to move beyond the hype and understand the mechanics of AI, read the full breakdown here: https://mytechnotalent.substack.com
#ArtificialIntelligence #MachineLearning #DeepLearning #ReverseEngineering #Engineering #BuildInPublic
In the Reverse Engineering world, we have a rule: You don't own it until you can take it apart.
The same applies to Artificial Intelligence.
We are currently drowning in API wrappers. Everyone is "building AI apps," but very few people are looking at the wiring underneath. To truly understand modern LLMs, I decided to stop using libraries. I went back to the drawing board to build a custom architecture from scratch.
Meet SARAN (Shallow Auto-Regressive Attention Network).
It’s not designed to beat GPT-4. It’s designed to be transparent. 🔹 I built a strict 15-stage computational graph. 🔹 I manually implemented backpropagation to trace the gradients. 🔹 I scaled it to a 354M parameter model to watch how it learns.
I’ve documented the entire build log—including the architecture decisions and the "why" behind the math—in my new engineering newsletter, Bits & Neurons.
If you want to move beyond the hype and understand the mechanics of AI, read the full breakdown here: https://mytechnotalent.substack.com
#ArtificialIntelligence #MachineLearning #DeepLearning #ReverseEngineering #Engineering #BuildInPublic
CharlesNepote
boosted
Je pense que si nous arrêtions de les appeler « intelligences artificielles » pour « traitements statistiques de la langue », cela redistribuerait les cartes. #ia #ai #ChatGPT #copilot #Statistics #MachineLearning https://www.lemonde.fr/idees/article/2026/01/25/entre-grokipedia-et-wikipedia-il-s-agit-de-determiner-quelles-conceptions-du-savoir-nous-souhaitons-defendre_6664020_3232.html
devSJR :python: :rstats:
boosted
AI is stepping up to help diagnose cervical cancer more accurately. Studies show AI can detect cancer cells in biopsies with comparable results to human pathologists. But challenges like inconsistent training data and the need for clinical validation remain. It's a promising tool, but not quite ready to replace doctors yet. 😐
AI is stepping up to help diagnose cervical cancer more accurately. Studies show AI can detect cancer cells in biopsies with comparable results to human pathologists. But challenges like inconsistent training data and the need for clinical validation remain. It's a promising tool, but not quite ready to replace doctors yet. 😐
Je pense que si nous arrêtions de les appeler « intelligences artificielles » pour « traitements statistiques de la langue », cela redistribuerait les cartes. #ia #ai #ChatGPT #copilot #Statistics #MachineLearning https://www.lemonde.fr/idees/article/2026/01/25/entre-grokipedia-et-wikipedia-il-s-agit-de-determiner-quelles-conceptions-du-savoir-nous-souhaitons-defendre_6664020_3232.html
CLI for working with Apple Core ML models
https://github.com/schappim/coreml-cli
#HackerNews #CLI #AppleCoreML #CoreMLTools #MachineLearning #GitHub
Missing momentum is effectively the only way that the main detectors at the LHC (like CMS) can measure the indirect signs of neutrinos (or undiscovered particles). This #CMSPaper present how we use #machinelearning to make our missing momentum measurement much, much better: arxiv.org/abs/2509.12012
GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers
https://gptzero.me/news/neurips/
#HackerNews #GPTZero #NeurIPS2025 #Hallucinations #AIresearch #MachineLearning
Binary Fuse Filters: Fast and Smaller Than XOR Filters
https://arxiv.org/abs/2201.01174
#HackerNews #BinaryFuseFilters #FastAlgorithms #XORFilters #DataStructures #MachineLearning
Three types of LLM workloads and how to serve them
https://modal.com/llm-almanac/workloads
#HackerNews #LLMworkloads #AItechnology #MachineLearning #ModalServices #TechInsights
This #CMSPaper investigates different #AI #machinelearning methods that aim to find jets that are inconsistent with the standard model. It shows that a new method called #Wasserstein normalized autoencodes works much better than other neural networks arxiv.org/abs/2510.02168
Batmobile: 10-20x Faster CUDA Kernels for Equivariant Graph Neural Networks
https://elliotarledge.com/blog/batmobile
#HackerNews #Batmobile #Equivariant #Graph #Neural #Networks #CUDA #Kernels #MachineLearning #AI
Ulrike Hahn
boosted
📢 Applications open on 19 Jan for the 7-week #Mathematics #SummerSchool in London.
You will develop the maths skills and intuition necessary to enter the #TheoreticalNeuroscience & #MachineLearning field.
ℹ️ Find out more & register for the information webinar 👉 https://www.ucl.ac.uk/life-sciences/gatsby/study-and-work/gatsby-bridging-programme
