O čem to celé bude?

Have you heard about AI that is able to create images from text? Have you noticed how good the language translation services have become? Are you amazed that you can talk to your computer and it can understand your commands with shocking clarity? All this and more have resulted from the growing popularity of neural networks and deep learning.

In this workshop we will introduce you to the world of neural networks. We will show you their applications, introduce you to what is happening behind the scenes and show you how you can train your network. You don’t have time for that? No worries! We will also show how you can empower your data science toolbox with existing powerful networks others have already created.

What will you learn

A brief tour showing the NNs from various fields that made headlines

  • Vision (Detectron2, DeepFaceLab, …)
  • Text (GPL3)
  • Text to image and image to text conversion (DallE (OpenAI), StableDifussion)

Behind the scenes – basics of how neural networks work

  • The Perceptron (single neuron network) and its inspiration from biology
  • Multilayer Perceptron
  • XOR problem and how it is solved by multiple layers
  • How multiple layers make better learners

How to train neural networks

  • Backpropagation Algorithm and Gradient Descent
  • Brief overview of advanced optimization algorithms
  • Where to find data for training? (Kaggle, Google Dataset search)

Brief overview of other types of neural networks

  • Convolutional
  • Recurrent
  • Adversarial
  • Transformers

Standing on the shoulders of giants - using pretrained models

  • Why to use them?
  • Where to find them?
  • Guided activity with pretrained models (depending on time available)

Who this course is for

For women at the beginning of their data science journey, who already have knowledge of basics (attended Data Science Foundation course, or have been exposed to some other resources for beginners).

Speakers

Stanislav Smatana

Stanislav leads Anthology data science team Morpheus, which is focused on machine learning and statistical modeling. Stano works on data analysis, statistical modeling and machine learning model training. Apart from his job at Anthology, he is also PhD student at Faculty of Information Technology at Brno University of Technology, where he does research on analysis of biological molecular data.

Ján Bogár

Jan is a member of the Morpheus data science team in Anthology. Even though he originally studied physics, Jan is focused on data science and machine learning. During his career, he worked with various applications of machine learning – from automatic coughing counting in clinical trials to analysis of scanned historical documents.

Elisa Suarez

Eli is a data scientist based in Bogota, Colombia, working in Anthology's data science team Morpheus. Even though she studied Mechatronics at university, she specializes in computer vision and convolutional neural networks. She has worked in automatic counting for logistic applications and pose inference of objects for robotics using depth images.

Přihlásit se na termín:

23
.
11
.
29
.
11
.

Online

No items found.
17:00
21:00
Hlavní lektor
Kontaktní osoba