Gregor Kasieczka
@kasieczka.bsky.social
Full Professor for Machine Learning in Particle Physics at Universität Hamburg | Searching for new physics with #CMSExperiment | He/Him
Today we celebrate Erik Buhmann who just defended his excellent doctoral work on generative AI applied to detector simulation!
Congratulations Erik for this #EPiC [1] PhD!
[1] arxiv.org/abs/2301.08128
Congratulations Erik for this #EPiC [1] PhD!
[1] arxiv.org/abs/2301.08128
January 29, 2025 at 7:24 PM
Today we celebrate Erik Buhmann who just defended his excellent doctoral work on generative AI applied to detector simulation!
Congratulations Erik for this #EPiC [1] PhD!
[1] arxiv.org/abs/2301.08128
Congratulations Erik for this #EPiC [1] PhD!
[1] arxiv.org/abs/2301.08128
🎉🎉🎉 Many congratulations to Manuel for defending his PhD thesis on detecting anomalies in particle physics data!
He did a great job, not only on the CATHODE approach for anomaly detection based on weak supervision, but also applying it to data collected by the CMS experiment at CERN.
#PhDone
He did a great job, not only on the CATHODE approach for anomaly detection based on weak supervision, but also applying it to data collected by the CMS experiment at CERN.
#PhDone
January 28, 2025 at 3:45 PM
🎉🎉🎉 Many congratulations to Manuel for defending his PhD thesis on detecting anomalies in particle physics data!
He did a great job, not only on the CATHODE approach for anomaly detection based on weak supervision, but also applying it to data collected by the CMS experiment at CERN.
#PhDone
He did a great job, not only on the CATHODE approach for anomaly detection based on weak supervision, but also applying it to data collected by the CMS experiment at CERN.
#PhDone
We show that a hybrid (conserving+breaking) architecture combines the advantages of symmetry-conserving networks (fast learning) and breaking ones (higher maximum performance).
Great work by Seth with @aishikghosh.bsky.social, Ed, @danielwhiteson.bsky.social
Full paper: arxiv.org/abs/2412.18773
Great work by Seth with @aishikghosh.bsky.social, Ed, @danielwhiteson.bsky.social
Full paper: arxiv.org/abs/2412.18773
January 6, 2025 at 8:44 AM
We show that a hybrid (conserving+breaking) architecture combines the advantages of symmetry-conserving networks (fast learning) and breaking ones (higher maximum performance).
Great work by Seth with @aishikghosh.bsky.social, Ed, @danielwhiteson.bsky.social
Full paper: arxiv.org/abs/2412.18773
Great work by Seth with @aishikghosh.bsky.social, Ed, @danielwhiteson.bsky.social
Full paper: arxiv.org/abs/2412.18773
Realised with the very nice talk by Marie yesterday that I had never posted about our joint paper: "Accurate and robust methods for direct background estimation in resonant anomaly detection"
Check out arxiv.org/abs/2411.00085 for a new idea on how to extract anomalies!
Check out arxiv.org/abs/2411.00085 for a new idea on how to extract anomalies!
December 18, 2024 at 7:37 AM
Realised with the very nice talk by Marie yesterday that I had never posted about our joint paper: "Accurate and robust methods for direct background estimation in resonant anomaly detection"
Check out arxiv.org/abs/2411.00085 for a new idea on how to extract anomalies!
Check out arxiv.org/abs/2411.00085 for a new idea on how to extract anomalies!
New paper on #arxiv, inspired by discussions this summer at the Aspen Center for Physics
🤔 Foundation models need huge amounts of training data.
🤔 There now exist large volumes of #OpenData by e.g. @cmsexperiment.bsky.social
💡 Why not use that data for training?
🤔 Foundation models need huge amounts of training data.
🤔 There now exist large volumes of #OpenData by e.g. @cmsexperiment.bsky.social
💡 Why not use that data for training?
December 17, 2024 at 7:21 AM
New paper on #arxiv, inspired by discussions this summer at the Aspen Center for Physics
🤔 Foundation models need huge amounts of training data.
🤔 There now exist large volumes of #OpenData by e.g. @cmsexperiment.bsky.social
💡 Why not use that data for training?
🤔 Foundation models need huge amounts of training data.
🤔 There now exist large volumes of #OpenData by e.g. @cmsexperiment.bsky.social
💡 Why not use that data for training?
Greetings from our AI & Physics with friends workshop, this year hosted by Markus @kitkarlsruhe.bsky.social!
Besides great talks by Hamburg & other students, we also got to see the local KATRIN experiment which is trying to measure the neutrino mass
Besides great talks by Hamburg & other students, we also got to see the local KATRIN experiment which is trying to measure the neutrino mass
December 16, 2024 at 4:56 PM
Greetings from our AI & Physics with friends workshop, this year hosted by Markus @kitkarlsruhe.bsky.social!
Besides great talks by Hamburg & other students, we also got to see the local KATRIN experiment which is trying to measure the neutrino mass
Besides great talks by Hamburg & other students, we also got to see the local KATRIN experiment which is trying to measure the neutrino mass
Congratulations to Sebastian Bieringer who successfully defended his PhD today!
This was a *truly* interdisciplinary work [1] between particle physics, maths/statistics (thanks to excellent co-supervisor Mathias Trabs) and machine learning!
[1] inspirehep.net/authors/1906...
This was a *truly* interdisciplinary work [1] between particle physics, maths/statistics (thanks to excellent co-supervisor Mathias Trabs) and machine learning!
[1] inspirehep.net/authors/1906...
December 10, 2024 at 2:51 PM
Congratulations to Sebastian Bieringer who successfully defended his PhD today!
This was a *truly* interdisciplinary work [1] between particle physics, maths/statistics (thanks to excellent co-supervisor Mathias Trabs) and machine learning!
[1] inspirehep.net/authors/1906...
This was a *truly* interdisciplinary work [1] between particle physics, maths/statistics (thanks to excellent co-supervisor Mathias Trabs) and machine learning!
[1] inspirehep.net/authors/1906...
Finally on #arxiv: Massive @cmsexperiment.bsky.social paper searching for anomalies in data: arxiv.org/abs/2412.03747
No new physics but it shows that ML-based anomaly detection methods can indeed be applied to collider data and offer a broad sensitivity to different potential signals!
No new physics but it shows that ML-based anomaly detection methods can indeed be applied to collider data and offer a broad sensitivity to different potential signals!
December 10, 2024 at 7:30 AM
Finally on #arxiv: Massive @cmsexperiment.bsky.social paper searching for anomalies in data: arxiv.org/abs/2412.03747
No new physics but it shows that ML-based anomaly detection methods can indeed be applied to collider data and offer a broad sensitivity to different potential signals!
No new physics but it shows that ML-based anomaly detection methods can indeed be applied to collider data and offer a broad sensitivity to different potential signals!
Got interviewed by German national TV today about the very inspired hashtag#NobelPrize decision.
Concepts from physics have been pivotal in developing machine learning & in turn we can use modern AI tools to do amazing physics.
Video here (in German): www.tagesschau.de/multimedia/v...
Concepts from physics have been pivotal in developing machine learning & in turn we can use modern AI tools to do amazing physics.
Video here (in German): www.tagesschau.de/multimedia/v...
October 8, 2024 at 3:08 PM
Got interviewed by German national TV today about the very inspired hashtag#NobelPrize decision.
Concepts from physics have been pivotal in developing machine learning & in turn we can use modern AI tools to do amazing physics.
Video here (in German): www.tagesschau.de/multimedia/v...
Concepts from physics have been pivotal in developing machine learning & in turn we can use modern AI tools to do amazing physics.
Video here (in German): www.tagesschau.de/multimedia/v...
Now at #DPG24: Our postdoc #LisaBenato with an awesome overview of searches for long-lived particles! Great to see over the last years the path from wild ideas to published results by @cmsexperiment.bsky.social @atlasexperiment.bsky.social @lhcb.bsky.social & more
indico.desy.de/event/42884/...
indico.desy.de/event/42884/...
March 7, 2024 at 10:42 AM
Now at #DPG24: Our postdoc #LisaBenato with an awesome overview of searches for long-lived particles! Great to see over the last years the path from wild ideas to published results by @cmsexperiment.bsky.social @atlasexperiment.bsky.social @lhcb.bsky.social & more
indico.desy.de/event/42884/...
indico.desy.de/event/42884/...
Had a great time teaching generative models at the COFI school in Puerto Rico
Don't usually post too many photos, but this might be a justified exception
Program & slides at: indico.cern.ch/event/1299889/
With many thanks to the organisers Kevin, Mayda, and Michael!
Don't usually post too many photos, but this might be a justified exception
Program & slides at: indico.cern.ch/event/1299889/
With many thanks to the organisers Kevin, Mayda, and Michael!
December 15, 2023 at 3:10 PM
Had a great time teaching generative models at the COFI school in Puerto Rico
Don't usually post too many photos, but this might be a justified exception
Program & slides at: indico.cern.ch/event/1299889/
With many thanks to the organisers Kevin, Mayda, and Michael!
Don't usually post too many photos, but this might be a justified exception
Program & slides at: indico.cern.ch/event/1299889/
With many thanks to the organisers Kevin, Mayda, and Michael!
Congratulations to Joerg who just successfully defended his PhD searching for very far-out particles 🎉🎉🎉
In case you want to know more, @cmsexperiment.bsky.social has you covered: cms.cern/news/searchi...
#PhDdone
In case you want to know more, @cmsexperiment.bsky.social has you covered: cms.cern/news/searchi...
#PhDdone
December 5, 2023 at 1:41 PM
Congratulations to Joerg who just successfully defended his PhD searching for very far-out particles 🎉🎉🎉
In case you want to know more, @cmsexperiment.bsky.social has you covered: cms.cern/news/searchi...
#PhDdone
In case you want to know more, @cmsexperiment.bsky.social has you covered: cms.cern/news/searchi...
#PhDdone
Not a geologist, but I think it's beautiful to see stalactites and stalagmites growing towards each other
December 5, 2023 at 9:26 AM
Not a geologist, but I think it's beautiful to see stalactites and stalagmites growing towards each other
🚨New paper alert🚨
In arxiv.org/abs/2312.00123 we show how to scale generative models to more complex data, and learn many properties (1st pic) beyond kinematics for 10 types of jets (2nd pic) based on the JetClass dataset
Code to compare, etc soon at: github.com/uhh-pd-ml/be...
In arxiv.org/abs/2312.00123 we show how to scale generative models to more complex data, and learn many properties (1st pic) beyond kinematics for 10 types of jets (2nd pic) based on the JetClass dataset
Code to compare, etc soon at: github.com/uhh-pd-ml/be...
December 4, 2023 at 9:52 AM
🚨New paper alert🚨
In arxiv.org/abs/2312.00123 we show how to scale generative models to more complex data, and learn many properties (1st pic) beyond kinematics for 10 types of jets (2nd pic) based on the JetClass dataset
Code to compare, etc soon at: github.com/uhh-pd-ml/be...
In arxiv.org/abs/2312.00123 we show how to scale generative models to more complex data, and learn many properties (1st pic) beyond kinematics for 10 types of jets (2nd pic) based on the JetClass dataset
Code to compare, etc soon at: github.com/uhh-pd-ml/be...
Kicking off the final morning at the #AISSAI workshop is Tommaso Dorigo with a great overview of techniques to handle uncertainties in HEP data analysis. See his slides for details and many further pointers:
indico.in2p3.fr/event/30589/...
indico.in2p3.fr/event/30589/...
December 1, 2023 at 8:28 AM
Kicking off the final morning at the #AISSAI workshop is Tommaso Dorigo with a great overview of techniques to handle uncertainties in HEP data analysis. See his slides for details and many further pointers:
indico.in2p3.fr/event/30589/...
indico.in2p3.fr/event/30589/...
Best door sign design ever at guest offices of the also otherwise brilliant AI uncertainty workshop.
Full program and slides for all talks at: indico.in2p3.fr/event/30589/...
Thanks to DavidR #AISSAI et al for great organisation
Full program and slides for all talks at: indico.in2p3.fr/event/30589/...
Thanks to DavidR #AISSAI et al for great organisation
November 30, 2023 at 2:02 PM
Best door sign design ever at guest offices of the also otherwise brilliant AI uncertainty workshop.
Full program and slides for all talks at: indico.in2p3.fr/event/30589/...
Thanks to DavidR #AISSAI et al for great organisation
Full program and slides for all talks at: indico.in2p3.fr/event/30589/...
Thanks to DavidR #AISSAI et al for great organisation
Greetings from the last big meeting of our cluster of excellence quantum universe for this year
Caught in discussion with our excellent spokesperson Erika
(Luckily the top-secret sheet of paper is sufficiently blurry)
Caught in discussion with our excellent spokesperson Erika
(Luckily the top-secret sheet of paper is sufficiently blurry)
November 21, 2023 at 11:34 AM
Greetings from the last big meeting of our cluster of excellence quantum universe for this year
Caught in discussion with our excellent spokesperson Erika
(Luckily the top-secret sheet of paper is sufficiently blurry)
Caught in discussion with our excellent spokesperson Erika
(Luckily the top-secret sheet of paper is sufficiently blurry)
Do you want to know what @cmsexperiment.bsky.social is up to in terms of #anomaly detection?
Look no further than our note out now! cds.cern.ch/record/28810...
While no results on data (yet), it gives a shape of the things to come...
Look no further than our note out now! cds.cern.ch/record/28810...
While no results on data (yet), it gives a shape of the things to come...
November 20, 2023 at 4:33 PM
Do you want to know what @cmsexperiment.bsky.social is up to in terms of #anomaly detection?
Look no further than our note out now! cds.cern.ch/record/28810...
While no results on data (yet), it gives a shape of the things to come...
Look no further than our note out now! cds.cern.ch/record/28810...
While no results on data (yet), it gives a shape of the things to come...
Sorry for silence recently, busy hosting #ML4Jets in Hamburg.
Thanks for all the seriously excellent developments on AI applied to particle physics (& beyond)
Not going to summarize the 133 (!) contributions here, but you can check out the slides: indico.cern.ch/event/1253794
Thanks for all the seriously excellent developments on AI applied to particle physics (& beyond)
Not going to summarize the 133 (!) contributions here, but you can check out the slides: indico.cern.ch/event/1253794
November 10, 2023 at 12:37 PM
Sorry for silence recently, busy hosting #ML4Jets in Hamburg.
Thanks for all the seriously excellent developments on AI applied to particle physics (& beyond)
Not going to summarize the 133 (!) contributions here, but you can check out the slides: indico.cern.ch/event/1253794
Thanks for all the seriously excellent developments on AI applied to particle physics (& beyond)
Not going to summarize the 133 (!) contributions here, but you can check out the slides: indico.cern.ch/event/1253794
Application for the #AspenPhysics summer workshops is now open at [1].
If you'd like to join our workshop on 'Fundamental Physics in the Era of Big Data and Machine Learning' you know where to go..
[1] aspenphys.org/physicists/s...
If you'd like to join our workshop on 'Fundamental Physics in the Era of Big Data and Machine Learning' you know where to go..
[1] aspenphys.org/physicists/s...
October 19, 2023 at 12:00 PM
Application for the #AspenPhysics summer workshops is now open at [1].
If you'd like to join our workshop on 'Fundamental Physics in the Era of Big Data and Machine Learning' you know where to go..
[1] aspenphys.org/physicists/s...
If you'd like to join our workshop on 'Fundamental Physics in the Era of Big Data and Machine Learning' you know where to go..
[1] aspenphys.org/physicists/s...
The physics gang problem is really getting out of hand in Hamburg.
October 15, 2023 at 8:54 AM
The physics gang problem is really getting out of hand in Hamburg.
New #paper on arXiv: 'Full Phase Space Resonant Anomaly Detection' (arxiv.org/abs/2310.06897)
A nice side effect of recent better generative models for HEP data is that we can now learn and interpolate in much higher dimensions, improving anomaly detection
A nice side effect of recent better generative models for HEP data is that we can now learn and interpolate in much higher dimensions, improving anomaly detection
October 12, 2023 at 6:52 AM
New #paper on arXiv: 'Full Phase Space Resonant Anomaly Detection' (arxiv.org/abs/2310.06897)
A nice side effect of recent better generative models for HEP data is that we can now learn and interpolate in much higher dimensions, improving anomaly detection
A nice side effect of recent better generative models for HEP data is that we can now learn and interpolate in much higher dimensions, improving anomaly detection