Building a Responsible AI strategy

FATE AI ( Fair, Accountable, Transparent, Ethical)
What is Responsible AI? What are the dangers? Where do you find the balance? How do I build a Responsible AI strategy for my organisation? And what tools are already there to assist you?
Come join this session and get answers on all these questions?

Explaining your AI models - Why & How

In a world where more and more AI models are used by big and small companies. It is also important to be able to explain why certain decisions are made. To be sure your model is not biased by gender or ethnicity.
In this session we have a look at how Azure Machine Learning services can help you to understand your machine learning model and how to make sure your companies AI model is not biased.

A Bot - From Idea to Deployment

Bot Framework Composer does not have any build in version controls to be able to track your code changes.
In this session we will go through all the different steps needed to build a bot and deploy it on different environments via DevOps.

How to embed AI, when you have no AI skills

In a world where AI is getting more and more important. It is good there are tools that can help you to add advanced AI skills into your own application or even DWH without having to know any Machine or Deep Learning techniques. This session will show you around in the Azure AI Stack and explain how you can use them!

Building chat bots with Microsoft - 4 different flavors

In 2019 a lot has changed in the Microsoft Bot space. Where before there was the Microsoft Bot Framework SDK, LUIS and QnA Maker, the solution stack has been extended with Power Virtual Agent and the Bot Framework Composer. How easy are these services and what are their capabilities? And how do you choose the best fit for your project?

Introduction to Microsoft SEAL - Homomorphic encryption technology

In a world where privacy is the biggest concern but where data is the new oil. We need new kinds of encryption technologies where the data is still usable for computation.

This session will explain you some theory about homomorphic encryption. To help you understand what it is, what you can do with it, but most important why it can be so useful.
At the end of the session you will be able to use it in your own data application.

Dificult words but easily explained

Advanced Text Analytics with Cognitive Services and custom ML

In a world where knowledge is power we notice that more and more companies want to get more information out of text based documents.
Microsoft provides already some great cognitive services, that can deliver some nice results, but we want to go to the next level.
In this session I will show how to combine all these services and extend it with some custom Machine Learning models.

Technologies:
- Azure Notebooks
- Cognitive Services
- Data Factory
- Data Bricks

Coding Languages:
- Python

Help! I can't find what I need.

We live in a world full of unstructured data, with data that is not easy to search in.
Video/Audio/Images/PDF/...
This talk will explain you how you can manage this documents and give your users the possibility to search through them.
How can you extend Cognitive Search with more powerful tools? Tools like Cognitive Services and Custom Machine Learning can really boost the user experience. Lets see how!

Introduction to Bot Framework Composer

At Ignite, Microsoft announced a visual tool to build chatbots quickly. With features as language generation, language understanding, QnA, and many more with a low-code experience.
In this session, I will build a chatbot from scratch and showing many different features of Bot Framework Composer

Accelerate your time to get a production ready ML Model.

A first step to learn some Machine Learning is by making use of the Automated Machine Learning services of Azure.

It enables developers with little machine learning skills to build ML models with high scale, efficiency, and productivity all while sustaining model quality.

This demo based session will start with collecting the data up to deploying models. ML terminology will be explained during the session so people with no AI background can follow.

Learn how to train high accuracy machine learning models using automated machine learning

Intelligent experiences powered by AI can seem like magic to users. Developing them, however, is cumbersome involving a series of sequential and interconnected decisions along the way that are pretty time consuming. What if there was an automated service that identifies the best machine learning pipelines for a given problem/data? Automated machine learning does exactly that! Learn how to quickly and efficiently build high accuracy ml models for classification, regression and forecasting scenarios using code and code-free pathways.

Knowledge Mining with Azure Search (Whats New?)

Azure Search is a Microsoft service that exists already for a while. But many do not know powerful and how easy this service is to use. In this demo based session I will show all the different functionalities of this services but also how you can extend it.

How to extend the Power Virtual Agent with QnA

QnA maker is a great tool to easily add a FAQ to your chatbot. But by default there is not way to integrate it into the Virtual Agent. During this session we will have a look at the QnA Maker and will see how to integrate it with the Power Virtual Agent.

Building a great ML algorithm with 0 skills

Did you know you don't need to know a lot about AI to build your own Machine Learning Model?
With Microsoft Automated ML you have a service that can help you set up a whole build pipeline that can generate top notch models without you having to write any code. This demo based session will give you the in- en outs of this service.

Microsoft AI - From Zero to Hero

Many think they don't have the skills to implement AI in their organisation. But at the end its not that hard.
In this workshop we will start with the easy pre-build AI models that are available. Then we will go through some small theory. And then we will move on to building your own AI models in 3 different levels (easy - moderate - difficult).
- Automated Machine Learning - Reusing the knowledge of millions of experiments
- Machine Learning Studio - Drag & Drop your AI model together
- Jupiter Notebook - Write your model from scratch (or not?)