Third Machine Learning in High Energy Physics Summer School 2017
from
Monday 17 July 2017 (07:00)
to
Sunday 23 July 2017 (18:00)
Monday 17 July 2017
08:50
Opening words
Opening words
08:50 - 09:00
Room: Reading University
Contributions
08:50
'Welcome to MLHEP' Opening Words
-
Andrey Ustyuzhanin
(
Yandex School of Data Analysis (RU)
)
Ulrik Egede
(
Imperial College (GB)
)
09:00
Day 1 Lectures
Day 1 Lectures
09:00 - 10:30
Room: Reading University
Contributions
09:00
Lecture 1. Introduction. Why ML works. Overfitting. Model Selection. Figures of Merits. Linear Models. Regularization. Logistic Regression.
-
Alexey Artemov
(
Yandex
)
10:30
Coffee break
Coffee break
10:30 - 10:50
Room: Reading University
10:50
Competition Introduction
Competition Introduction
10:50 - 11:10
Room: Reading University
Contributions
10:50
Competition Introduction
-
Andrey Ustyuzhanin
(
Yandex School of Data Analysis (RU)
)
11:10
Day 1 Seminars
Day 1 Seminars
11:10 - 12:40
Room: Reading University
Contributions
11:10
Seminar 1. Python data crunching: numpy, root_numpy, pandas (very short), matplotlib.
-
Alexander Panin
(
Yandex School of Data Analysis (RU)
)
12:40
Lunch
Lunch
12:40 - 14:00
Room: Reading University
14:00
Day 1 Lectures
Day 1 Lectures
14:00 - 15:30
Room: Reading University
Contributions
14:00
Lecture 2. Decision Trees. Bagging. Ensembles. RandomForest. AdaBoost. GB-Reweighting. Gradient Boosting. Decorrelation of features with predictions.
-
Alexey Artemov
(
Yandex
)
15:30
Coffee break
Coffee break
15:30 - 15:50
Room: Reading University
15:50
Day 1 Seminars
Day 1 Seminars
15:50 - 17:20
Room: Reading University
Contributions
15:50
Seminar 2. Matplotlib, Trees & Linear Models in SciKit Learn. Overfitting checks & prevention.
-
Alexander Panin
(
Yandex School of Data Analysis (RU)
)
19:00
Welcome Reception
Welcome Reception
19:00 - 20:30
Room: Reading University
Tuesday 18 July 2017
09:00
Day 2 Lectures
Day 2 Lectures
09:00 - 10:30
Room: Reading University
Contributions
09:00
Lecture 3. Ensembling: Adaboost, Stacking. Gradient Boosting.
10:30
Coffee break
Coffee break
10:30 - 11:00
Room: Reading University
11:00
Day 2 Seminars
Day 2 Seminars
11:00 - 12:30
Room: Reading University
Contributions
11:00
Seminar 3. Ensembling: Adaboost, Stacking. Gradient Boosting.
12:30
Lunch
Lunch
12:30 - 13:30
Room: Reading University
13:30
Day 2 Lectures
Day 2 Lectures
13:30 - 15:00
Room: Reading University
Contributions
13:30
Lecture 4. Hyperparameters optimization techniques and methods. Feature selection. TPE. Gaussian Processes
15:00
Coffee break
Coffee break
15:00 - 15:30
Room: Reading University
15:30
Day 2 seminrs
Day 2 seminrs
15:30 - 17:00
Room: Reading University
Contributions
15:30
Seminar 4. Optimization of hyper-parameters, applications to challenge.
17:00
Break
Break
17:00 - 17:15
Room: Reading University
17:15
Mike Williams
Mike Williams
17:15 - 18:45
Room: Reading University
Wednesday 19 July 2017
09:00
09:00 - 21:00
Room: Reading University
Thursday 20 July 2017
09:00
Day 4 Lectures
Day 4 Lectures
09:00 - 10:30
Room: Reading University
Contributions
09:00
Lecture 5. Introduction into Neural Network models. Multi-layered Perceptron. Backpropagation.
10:30
Coffee break
Coffee break
10:30 - 11:00
Room: Reading University
11:00
Day 4 Seminars
Day 4 Seminars
11:00 - 12:30
Room: Reading University
Contributions
11:00
Seminar 5. Computing gradient by hand. Tensorflow. Keras. Remote machine setup. Getting familiar with GPU.
12:30
Lunch
Lunch
12:30 - 13:30
Room: Reading University
13:30
Day 4 Lectures
Day 4 Lectures
13:30 - 15:00
Room: Reading University
Contributions
13:30
Lecture 6. Intro into Deep learning. Convolutional Neural Network. Model zoo. Augmentation.
15:00
Coffee break
Coffee break
15:00 - 15:30
Room: Reading University
15:30
Day 4 Seminars
Day 4 Seminars
15:30 - 17:00
Room: Reading University
Contributions
15:30
Seminar 6. Image recognition using convolutional neural networks.
17:00
Break
Break
17:00 - 17:15
Room: Reading University
17:15
Guest Lecture
Guest Lecture
17:15 - 18:45
Room: Reading University
Contributions
17:15
Guest Lecture
-
Noel Dawe
(
University of Melbourne (AU)
)
Friday 21 July 2017
09:00
ML Projects & Problems @ HEP
ML Projects & Problems @ HEP
09:00 - 10:30
Room: Reading University
10:30
Coffee break
Coffee break
10:30 - 11:00
Room: Reading University
11:00
ML at industry
ML at industry
11:00 - 12:30
Room: Reading University
12:30
Lunch
Lunch
12:30 - 13:30
Room: Reading University
13:30
Day 5 Lectures
Day 5 Lectures
13:30 - 15:00
Room: Reading University
Contributions
13:30
Lecture 7. Dimensionality Reduction. PCA. LDA. LLE. TSNE. Autoencoders.
15:00
Coffee break
Coffee break
15:00 - 15:30
Room: Reading University
15:30
Day 5 Seminars
Day 5 Seminars
15:30 - 17:00
Room: Reading University
Contributions
15:30
Seminar 7. PCA, Autoencoders (faces)
17:00
Break
Break
17:00 - 17:15
Room: Reading University
17:15
Tim Head
Tim Head
17:15 - 18:45
Room: Reading University
19:00
Dinner
Dinner
19:00 - 21:00
Saturday 22 July 2017
09:00
Day 6 Lectures
Day 6 Lectures
09:00 - 10:30
Room: Reading University
Contributions
09:00
Lecture 8 Generative Adversarial Networks, Metric Learning.
10:30
Coffee break
Coffee break
10:30 - 11:00
Room: Reading University
11:00
Day 6 Seminars
Day 6 Seminars
11:00 - 12:30
Room: Reading University
Contributions
11:00
Seminar 8 Generative Adversarial Networks (physics)
12:30
Lunch
Lunch
12:30 - 13:30
Room: Reading University
13:30
Day 6 Lectures
Day 6 Lectures
13:30 - 16:00
Room: Reading University
Contributions
13:30
Lecture 9. RNNs.
15:30
Certificates, Free Discussion
Sunday 23 July 2017
09:00
Lecture 10. Why Deep Learning works. Tips & tweaks.
Lecture 10. Why Deep Learning works. Tips & tweaks.
09:00 - 10:30
Room: Reading University
10:30
Break
Break
10:30 - 11:00
Room: Reading University
11:00
Awards & Winning solution presentation
Awards & Winning solution presentation
11:00 - 12:00
Room: Reading University