BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Introduction to Machine Learning
DTSTART;VALUE=DATE-TIME:20200420T080000Z
DTEND;VALUE=DATE-TIME:20200420T160000Z
DTSTAMP;VALUE=DATE-TIME:20240704T115515Z
UID:indico-event-1@meetings.aip.de
DESCRIPTION:• The concepts of Neural Networks and Machine Learning. Weak
  and Strong AI. Artificial neuron model.\n\n• Typical neural networks wo
 rkflow: features\, layers\, tensors\, gradients\, loss function\, back pro
 pagation\n• Types of problems: Regression\, Classification\, Object dete
 ction\, Segmentation\, Super Resolution\, Image generation\n• The feasib
 ility of Machine Learning for science and research\n\n• Advantages of Py
 thon and PyTorch for Neural Networks development\n• Tensor objects in Py
 thon and PyTorch\, interoperation with NumPy\n• Obtaining tensor gradien
 ts in PyTorch with Autograd\n• The structure of neural network implement
 ation in PyTorch: definition\, loss function\, backprop\, weights update\n
 \n• Our first neural network in PyTorch\, by example of Binary Classific
 ation\n• Python environments: Linux and Windows\, pip\, Anaconda\, Jupyt
 er Notebook\n• Training and evaluation of Binary Classification in PyTor
 ch: “Cat or Dog?”\n\n• PyTorch optimization algorithms in the contex
 t of neural network training\n• The generic structure of optimization al
 gorithms: parameters\, optimization step\, funciton evaluation closure\n
 • The SGD algorithm explained\n• Convolutional neural networks for cla
 ssification problems\n\n• Deploying LeNet to handwritten digits recognit
 ion for ZIP codes\n• Deploying Mobilenetv2 imagenet classification\n\nht
 tps://meetings.aip.de/event/5/sessions/1/
LOCATION:AIP SH/Lecture Hall
URL:https://meetings.aip.de/event/5/sessions/1/
END:VEVENT
END:VCALENDAR