#higgsml search results

10 years ago #OTD, @bingxu_ and @tqchenml quietly announced the birth of @XGBoostProject on the @kaggle #higgsml competition forum. Since then, XGBoost has skyrocketed in popularity to become the top ML go-to tool! 1/4

... with experience from #higgsml and #trackml competitions ! #hepml

With a special Far-Universe hackathon to prototype a hopefully #neurips 2024 competition. A #higgsml competition on steroids, where participant need to handle uncertainties on the inputs and output: the expected answer is a confidence interval ! #hepml

Super travail de @Giles_C_Strong qui a réétudié avec les outils moderne du deep learning le challenge #higgsml #kaggle que j’avais organisé en 2014

For anyone following #PyHEP2021 on Friday 15:30 CET, I'll be doing a live tutorial for LUMIN, where we'll train some DNNs and attempt to get near the top of the leaderboard for the #Kaggle #HiggsML challenge. indi.to/WnHcZ

Just out! With @AnaderiRu we've summed up our experience organising scientific challenges/competition on @kaggle #codalab and more. With sections on #HiggsML flavour of physics, @trackmllhc , #lhco2020, challenge organisation tips, and open datasets arxiv.org/abs/2012.08520

Indeed, bootstrapping was used to estimate the probability that #1 #2 #3 of #higgsml challenge exchange ranks. (this was proposed by @balazskegl ). It did work.

There is the technique used by @GaborMelis to win #higgsml challenge, which is to train on A B C D E separately (for 5 fold CV), and then « average » A B C D test on E, etc for each fold. Turned out the « average » was better than full training. Probably not always true.

It all started with some « what if » slides at the end of a talk on #HiggsML #kaggle challenge in the first occurrence of @ctdwit in Berkeley in 2015.

A different kind of challenges where concrete collaboration between participant is built in the platform and the incentives, and it works. The HEP anomalies one (top plot) is actually a spin off from the #higgsml challenge. #hepml.

Should be noted the key role of #XGboost (@tqchenml !) and have #BDT tackle directly the 4-momenta (instead of handcrafted variables). Both emerged during our #higgsml #kaggle challenge. Took 4 years, because, eh!, everything takes longer #irl but it was worth it. #hepml

#XGBoost released first 2014 for our #HiggsML @kaggle challenge. Adoption took long with a first physics paper using it dec 2017 by @ATLASexperiment arxiv.org/abs/1712.08891 .Would be interesting to compare the other two for our blobish distributions, few categorical features

In preparation for the upcoming @trackmllhc challenge I finally got round to running on the #HiggsML #Kaggle challenge: amva4newphysics.wordpress.com/2018/03/21/hig…

Hi to Gabor on the other side of the table, who won our #higgsml challenge in 2014.

Quite interesting they quote our #higgs #higgsml challenge in their leading quote!

Looking forward to it, I being at the two ends : ml challenge organisation #higgsml soon #trackml, and bird counting #stoceps

@balazskegl @SaclayCDS transverse mass ? Similar to the ττ invariant mass feature in the #HiggsML @kaggle challenge. MET large variance

Excellent nevertheless #HiggsML #kaggle

.@demishassabis tells us deep mind just won #HiggsML @kaggle challenge winner @GaborMelis #DSLHC15

...and yes, this is XGBoost (popular BDT since #higgsML) now interfaced to TMVA through R ! #DSLHC15

Our 1st Bootcamp on #DataScience. Lots of bright minds, including @GaborMelis winner of #HiggsML.Thanks to @PROTO204

Our first data-science bootcamp à la "Piscine" #HiggsML @MINES_ParisTech @SaclayCDS #communitybuilding

High Energy Physics meets Machine Learning: Enter the #HiggsML Challenge higgsml.lal.in2p3.fr via @ATLASexperiment

Register by 8 Sept, make #HiggsML the most popular Kaggle Open Challenge. tinyurl.com/pvu3dr9 via @ATLASexperiment

For more info about the #HiggsML challenge, see #ATLAS news articles, atlas.ch/news/2014/are-… & atlas.ch/news/2015/atla…

.@SaclayCDS #DataScience bootcamp on the #HiggsML challenge at @PROTO204, backend #Juju #Azure #ubuntu #canonical

Winners and organizers of #HiggsML @kaggle challenge at #NIPS2014 @GaborMelis @TimSalimans @balazskegl

“@CERN: High Energymeets Machine Learning: Enter the #HiggsML Challenge higgsml.lal.in2p3.fr via @ATLASexperiment”

Physique des hautes énergies + apprentissage automatique = #HiggsML Challenge higgsml.lal.in2p3.fr @ATLASexperiment

Apparently even Cedric Villani talks about the #HiggsML challenge :) fb.me/78V7lkQYT

Attention clôture des inscriptions pour le #HiggsML challenge le 8 septembre: tinyurl.com/pvu3dr9 @ATLASexperiment

Gábor Melis presenting his winning approach during the #HiggsML Kaggle challenge

#HiggsML workshop and visit of the @Kaggle challenge winners at @CERN. @SaclayCDS

The #HiggsML cartoon doing tracking ! In Tobias Golling talk at #DSLHC15...

...and yes, this is XGBoost (popular BDT since #higgsML) now interfaced to TMVA through R ! #DSLHC15

Putting physicists, data scientists, and management scientists in a room to tackle the unknown @SaclayCDS #HiggsML

RT @CERN: High Energy Physics meets Machine Learning: Enter the #HiggsML Challenge higgsml.lal.in2p3.fr

Tobias Golling (@Unigenews) mentions the #HiggsML challenge as an interesting follow-up question. #DSLHC15

Tobias Golling (@Unigenews) mentions the #HiggsML challenge as an interesting follow-up question.

Nice. Also how adorable rt @CERN: High Energy Physics meets Machine Learning: Enter the #HiggsML Challenge