Machine Learning
from
Monday, 20 May 2019 (11:00)
to
Wednesday, 22 May 2019 (16:00)
Monday, 20 May 2019
11:00
Introduction
Introduction
11:00 - 11:30
Room: Hotel Cenobio dei Dogi
11:30
Lecture 1 - Overview on ML/DL in HEP
-
Amir Farbin
(
University of Texas at Arlington
)
Lecture 1 - Overview on ML/DL in HEP
Amir Farbin
(
University of Texas at Arlington
)
11:30 - 13:00
Room: Hotel Cenobio dei Dogi
This lecture aims at giving a broad overview of main concepts in ML in the context of HEP applications.
13:00
Lunch break
Lunch break
13:00 - 14:30
Room: Hotel Cenobio dei Dogi
14:30
Lecture 2 - Introduction to the ML landscape and methods
-
Daniele Bonacorsi
(
BO
)
Lecture 2 - Introduction to the ML landscape and methods
Daniele Bonacorsi
(
BO
)
14:30 - 16:00
Room: Hotel Cenobio dei Dogi
An introduction to the (general, not HEP specific) ML landscape is given.
16:00
Coffee break
Coffee break
16:00 - 16:30
Room: Hotel Cenobio dei Dogi
16:30
Hands-on session
Hands-on session
16:30 - 17:30
Room: Hotel Cenobio dei Dogi
17:30
Compute vision and applications
-
Ferrari A.
(
Argo Vision
)
Compute vision and applications
Ferrari A.
(
Argo Vision
)
17:30 - 18:15
Room: Hotel Cenobio dei Dogi
18:15
ML in the online data acquisition
-
Cristiano Fanelli
(
ISS
)
ML in the online data acquisition
Cristiano Fanelli
(
ISS
)
18:15 - 18:40
Room: Hotel Cenobio dei Dogi
18:40
ML for the EIC
-
Dmitry Romanov
(
Jefferson Lab
)
ML for the EIC
Dmitry Romanov
(
Jefferson Lab
)
18:40 - 19:05
Room: Hotel Cenobio dei Dogi
Tuesday, 21 May 2019
09:00
Lecture 3 - End-2-end projects discussion: regression and classification
-
Daniele Bonacorsi
(
BO
)
Lecture 3 - End-2-end projects discussion: regression and classification
Daniele Bonacorsi
(
BO
)
09:00 - 10:30
Room: Hotel Cenobio dei Dogi
The attendees will be guided through an example of regression techniques.
10:30
Coffee break
Coffee break
10:30 - 11:00
Room: Hotel Cenobio dei Dogi
11:00
Lecture 4 - Rapid walk-through of Deep Learning
-
Amir Farbin
(
University of Texas at Arlington
)
Lecture 4 - Rapid walk-through of Deep Learning
Amir Farbin
(
University of Texas at Arlington
)
11:00 - 12:30
Room: Hotel Cenobio dei Dogi
The attendees will be guided through concepts and methods in Deep Learning.
12:30
Lunch break
Lunch break
12:30 - 14:30
Room: Hotel Cenobio dei Dogi
14:30
Hands-on session (cont'd)
Hands-on session (cont'd)
14:30 - 15:30
Room: Hotel Cenobio dei Dogi
15:30
Coffee break
Coffee break
15:30 - 16:00
Room: Hotel Cenobio dei Dogi
16:00
Hands-on session (cont'd)
Hands-on session (cont'd)
16:00 - 17:00
Room: Hotel Cenobio dei Dogi
17:00
Graph Network for High Energy Physics
-
Jean-Roch Vlimant
(
Caltech
)
Graph Network for High Energy Physics
Jean-Roch Vlimant
(
Caltech
)
17:00 - 17:45
Room: Hotel Cenobio dei Dogi
17:45
Improving Reaction Identification with ROOT TMVA
-
Derek Glazier
(
University of Edinburgh
)
Improving Reaction Identification with ROOT TMVA
Derek Glazier
(
University of Edinburgh
)
17:45 - 18:30
Room: Hotel Cenobio dei Dogi
Wednesday, 22 May 2019
09:00
Lecture 5 - Training models and Gradient Descent (bonus: Decision Trees)
-
Daniele Bonacorsi
(
BO
)
Lecture 5 - Training models and Gradient Descent (bonus: Decision Trees)
Daniele Bonacorsi
(
BO
)
09:00 - 10:30
Room: Hotel Cenobio dei Dogi
The attendees will be guided through what's under the hood in terms of training models. An additional lecture on Decision Trees is posted.
10:30
Coffee break
Coffee break
10:30 - 11:00
Room: Hotel Cenobio dei Dogi
11:00
Lecture 6 - Keras and/or Deep Learning in HEP
-
Amir Farbin
(
University of Texas at Arlington
)
Lecture 6 - Keras and/or Deep Learning in HEP
Amir Farbin
(
University of Texas at Arlington
)
11:00 - 12:30
Room: Hotel Cenobio dei Dogi
The attendes opted to be given a guided walk-through of the basics of Keras as from keras.io. Alternative choice was a seminar on Deep Learning in HEP, whose slides are anyway posted.
12:30
Lunch break
Lunch break
12:30 - 14:30
Room: Hotel Cenobio dei Dogi
14:30
Closeout
Closeout
14:30 - 16:00
Room: Hotel Cenobio dei Dogi
The closeout includes speakers presenting references and post-course self-training suggestions, as well as a discussion of attendees' open projects.