MLHEP 2015 summer school

from Thursday, 27 August 2015 (08:00) to Sunday, 30 August 2015 (23:00)
Academic University

        : Sessions
    /     : Talks
        : Breaks
27 Aug 2015
28 Aug 2015
29 Aug 2015
30 Aug 2015
AM
08:00
Registration (until 08:15) ()
08:15 --- breakfast ---
09:00
Welcome - Andrey Ustyuzhanin (Yandex School of Data Analysis (RU)) (until 09:35) (204)
09:00 Welcome to MLHEP   (204)
09:20 MLHEP Kaggle competition - Nikita Kazeev (Yandex School of Data Analysis (RU))   (204)
09:35 --- break ---
09:40
Advanced track -Dr Victor Kitov (MSU) (until 12:40) (206)
09:40 Reminder about major algorithms. Advanced aspects of their use. Model evaluation.   (206)
11:00 --- coffee break ---
11:20 Feature selection. Regularization.   (206)
09:40
Introductory track - Alexey Rogozhnikov (until 12:40) (204)
09:40 Intro: General pipeline, ML at a glance   (204)
11:00 --- coffee break ---
11:20 Intro: kNN, basic overfitting, roc curve, logistic regression   (204)
08:15 --- breakfast ---
09:30
Advanced track -Dr Victor Kitov (MSU) (until 12:30) (206)
09:30 model ensembling #1   (206)
10:50 --- coffee break ---
11:10 model ensembling #2   (206)
09:30
Introductory track - Alexey Rogozhnikov (until 12:30) (204)
09:30 neural networks   (204)
10:50 --- coffee break ---
11:10 decision tree, regression tree   (204)
08:15 --- breakfast ---
09:30
Advanced track -Dr Victor Kitov (MSU) (until 12:30) (206)
09:30 Linear dimensionality reduction.   (206)
10:50 --- coffee break ---
11:10 Non-linear dimensionality reduction. Kernel trick. Common kernels.   (206)
09:30
Introductory track - Alexey Rogozhnikov (until 12:30) (204)
09:30 Ensembles   (204)
10:50 --- coffee break ---
11:10 Gradient boosting; boosting to uniformity.   (204)
08:15 --- breakfast ---
09:00
Advanced track -Dr Victor Kitov (MSU) (until 12:00) (206)
09:00 Kernelized algorithms: SVM, regression, K-NN, PCA and their properties.   (206)
10:20 --- coffee break ---
10:40 Deep learning. Different kinds of learning.   (206)
09:00
Introductory track - Alexey Rogozhnikov (until 12:00) (204)
09:00 GBDT tuning, reweighting, testing hypotheses, gaussian processes for optimization   (204)
10:20 --- coffee break ---
10:40 Unsupervised learning, Latent Dirichlet Allocation, PCA. Fast predictions and pruning   (204)
PM
12:40 --- lunch ---
13:45
Advanced track - Tatiana Likhomanenko (National Research Centre Kurchatov Institute (RU)) (until 17:10) (206)
13:45 MLHEP introduction. ML vs MLHEP: correlation and agreement restrictions - Tatiana Likhomanenko (National Research Centre Kurchatov Institute (RU))   (206)
15:15 --- coffee break ---
15:40 sPlot technique. ML on sPlot data.   (206)
13:45
Introductory track - Nikita Kazeev (Yandex School of Data Analysis (RU)) (until 17:10) (204)
13:45 Introduction, course technicalities   (204)
15:15 --- coffee break ---
15:40 NumPy, Pandas, Matplotlib   (204)
17:10 --- break ---
17:20 Applying Machine Learning to LHC triggers optimization - Mika Anton Vesterinen (Ruprecht-Karls-Universitaet Heidelberg (DE))   (204)
19:00 --- Welcome dinner ---
12:30 --- lunch ---
13:45
Advanced track - Tatiana Likhomanenko (National Research Centre Kurchatov Institute (RU)) (until 17:10) (206)
13:45 Brave new boosting world: flatness boosting   (206)
15:15 --- coffee break ---
15:40 Brave new boosting world: reweighing boosting. Iterative learning   (206)
13:45
Introductory track - Nikita Kazeev (Yandex School of Data Analysis (RU)) (until 17:10) (204)
13:45 Overfitting, Cross-validation   (204)
15:15 --- coffee break ---
15:40 Sklearn: KNN, decision tree, logistic regression   (204)
17:10 --- break ---
17:30 Challenges of searching for physics signatures with a large amount of background - Patrick Haworth Owen (Imperial College Sci., Tech. & Med. (GB))   (204)
12:30 --- lunch ---
13:45
Advanced track - Tatiana Likhomanenko (National Research Centre Kurchatov Institute (RU)) (until 17:10) (206)
13:45 Meta Algorithms Applications in HEP.   (206)
15:15 --- coffee break ---
15:40 Hypotheses testing.   (206)
13:45
Introductory track - Nikita Kazeev (Yandex School of Data Analysis (RU)) (until 17:10) (204)
13:45 Ensemble algorithms: random forest, gradient boosting   (204)
15:15 --- coffee break ---
15:40 Hyperparam Optimization   (204)
17:10 --- break ---
17:30 MVA vs cut-and-count techniques in the SM model processes searches and measurements - Lesya Shchutska (University of Florida (US))   (204)
12:00 --- lunch ---
13:00
Advanced track - Tatiana Likhomanenko (National Research Centre Kurchatov Institute (RU)) (until 16:15) (206)
13:00 Summary: HEP ML analysis sketch, from data to discovery.   (206)
14:30 --- coffee break ---
14:45 My own first NN using theano.   (206)
13:00
Introductory track - Nikita Kazeev (Yandex School of Data Analysis (RU)) (until 16:15) (204)
13:00 PCA, RBM and alike on MNIST   (204)
14:30 --- coffee break ---
14:45 Clustering   (204)
16:15 --- break ---
16:20 Multivariate techniques in the Higgs search - Josh Bendavid (CERN)   (204)
18:00 --- transfer to the boat ---
19:00 --- channel cruise + closing dinner + awards ---
22:00 --- transfer to the hotel ---