Tweets
- Tweets
- Tweets & replies
- Photos & videos
@MLWave is blocked
Are you sure you want to view these Tweets? Viewing Tweets won't unblock @MLWave.
How I won InClass Kaggle competition "Predict Repeat Restaurant Bookings" by NTHU Taiwan. #vowpalwabbit #xgboost http://inclass.kaggle.com/c/predict-repeat-restaurant-bookings/forums/t/11340/congratulations-to-the-winner-and-top-entries/61829#post61829 …
Polynomial Learning (used by 3rd place Criteo CTR contest) https://github.com/JohnLangford/vowpal_wabbit/wiki/poly.pdf … Hogwild! Mode by @PaulMineiro https://github.com/JohnLangford/vowpal_wabbit/wiki/lrq_hogwild.pdf …
New VW Tutorial by John Langford for 7.8: Intro + reductions + log_multi + exploration library + AzureML https://github.com/JohnLangford/vowpal_wabbit/wiki/intro_2014.pdf …
11 retweets 21 favoritesUnsupervised Deep Learning by looking at semantic similarity between consecutive Youtube video frames http://arxiv.org/pdf/1412.6056.pdf … #yeslecan
Elastic Averaging SGD http://arxiv.org/abs/1412.6651 Faster convergence rate than Downpour SGD http://research.google.com/archive/large_deep_networks_nips2012.html …
1 favoriteThx @benhamner for showing the public Higgs leaderboard and not the private leaderboard https://www.youtube.com/watch?v=9Zag7uhjdYo#t=405 … :). @abhi1thakur @glouppe
YouTube
Contributing to a paper! Chapter: State-of-the-art results on Head Pose Estimation Dataset with Stacked Generalization and Random Forests.
"Fractional Max-Pooling" by Dr. Ben Graham (winner of CIFAR-10 on Kaggle) keeps improving. http://arxiv.org/abs/1412.6071
1 retweetConvolutional Nets and CIFAR-10: An Interview with Yann LeCun @ylecun http://blog.kaggle.com/2014/12/22/convolutional-nets-and-cifar-10-an-interview-with-yan-lecun/ …
Sub-Linear Debugging: I have a post on sub-linear debugging on Microsoft's machine learning blog.Online learni... http://bit.ly/1unXLHT
5 retweets 6 favorites
Amazing Kaggle solution to the @criteo challenge. 75 lines of code, no @scikit_learn https://www.kaggle.com/c/criteo-display-ad-challenge/forums/t/10322/beat-the-benchmark-with-less-then-200mb-of-memory …
Accelerating Random Forests in Scikit-Learn. Slides available at http://www.slideshare.net/glouppe/accelerating-random-forests-in-scikitlearn … @EuroSciPy
SlideShare
MT @freakonometrics: "10 R Packages to Win #Kaggle Competitions" http://fr.slideshare.net/DataRobot/final-10-r-xc-36610234 … #machinelearning #rstats
Lessons learned from the Hunt for Prohibited Content in @avito_ru 's advertisements on @kaggle (8th place, ty team!) http://mlwave.com/lessons-from-avito-prohibited-content-kaggle/ …
1 month left in the Higgs Boson ML Challenge: https://www.kaggle.com/c/higgs-boson That's long time in life of a tau particle, but not a data scientist!
12 retweets 15 favorites
That's a very cute name for an algorithm “@TwitterEng: Distributing the all-pairs similarity problem, with DIMSUM. https://blog.twitter.com/2014/all-pairs-similarity-via-dimsum …”
A look at the data science start-ups in the latest @yCombinator batch. http://mlwave.com/ycombinator-2014-data-science-start-ups/ … pic.twitter.com/bo7Msx5RBF
PhD candidate @UniofYork (Artificial Intelligence group); MSc. in #Physics; Former Quant +14 years; #ML #NLP #Quant
Twitter may be over capacity or experiencing a momentary hiccup.
Visit Twitter Status for more information.