KRZYSZTOF KUDRYŃSKI

DEEP LEARNING WORKSHOPS
***
CONFERENCE TALKS

On a daily basis, I am a designer of engineering solutions for artificial intelligence, autonomous vehicles, computer vision and robotics.

Everyday involvment into engineering topics is my main inspiration to prepare conference talks and practical Deep Learning workshops

Home: Welcome

Conference talks

From Backpropagation to Enterprise Deep Learning

From scientific journals and under-desk workstations, AI is entering the world of large corporations today, replacing traditional solutions with self-optimizing models.
During this talk, we will delve into the fundamentals of artificial intelligence by exploring backpropagation - the core mechanism powering AI systems. Through a hands-on coding session, we will develop a deep understanding of this concept by teaching a machine to comprehend music to finally embrace the challenges faced by framework designers and the biggest players in the market.
Join me to gain a fundamental understanding and valuable insights into this rapidly-evolving field.

Deep Learning at the Speed Of Light

What can your software do? Can it understand what it is looking at? Can it name what it is listening to? Does it have the ability to make its own choices, analyse new relations or create new content?

Today all of the above are becoming possible and the big players spare no pain to bring these emerging technologies into the enterprise world. In this presentation, we will go through the basics of Deep Learning to quickly realise the complexity of this industry. How to survive and succeed in such an environment as a company? What skills are required of the engineering teams? How Speed Of Light is the only stable reference? Join this session to answer these questions and have some fun going through real-life examples.

Traditional and Deep Learning based Vehicle Autonomy

Cars drive by themselves, drones deliver goods on their own, robots move around homes and offices. This is the reality we are entering with unprecedented pace.

In this talk I will go through the core algorithmic concepts behind vehicle autonomy as well as the most important Deep Learning architectures, gradually replacing traditional approaches. We will see how the components of autonomous vehicles are implemented and we will see how it works in practice.

Join me if you want to:

- understand how it works,
- learn how its implemented,
- list engineering skills most demanded in this field,
- just have some fun enjoying the demo footages.

Deep Learning to Fly

What does it take to make the machine understand its environment while moving around (or flying)? How to teach a drone to race on its own safely in unfriendly environment?

During our talk, you will participate in all the stages of this exciting adventure from the first lines of code, through algorithmic struggles up to complete system ready to participate in the drone racing league.

Deep Learning in Computer Vision

When we look at the world our brain instantly turns the images we see into information, intuition and feeling. Thanks to advances, both in hardware and algorithmic approach, we are making machines understand images they see.

In this talk you will learn how to design and develop such a network. We will go from the basic machine learning concepts, through the advanced practical tips, up to using the most-recent state-of-the art architectures in practice. You will see examples on how insanely complicated problems can be solved using a portable computer. And in the meanwhile you will understand what is going on, why new layers are added and why new concept have to be introduced.

Home: Tips & Advice

Workshops

Home: Services

Deep Learning with Convolutional Neural Networks: Understand, Develop and Apply

There seems to be a common agreement to treat Deep Learning as a magic blackbox. Developers only need to run `build`, `train` and `apply` methods and the frameworks will do the rest. However, frameworks are much more effective when in hands of a conscious developper. During this workshop we will use all the goodness of frameworks (eg. TensorFlow, Keras) but we will also learn how it works, why it works or if it fails - how to fix it.

During this workshop you will:
- gain deep understanding of Convolutional Neural Network - the most common Deep Learning tool at the moment.
- learn how to consciously design it so that it solves your more and more complicated problems
- see how you can reuse publicly available solutions, how to modify them, tune them to your own needs
- learn how to use frameworks like: TensorFlow and Keras to build Deep Learning solutions, monitor their training, and apply them in production.

Basic knowledge of Python is required.