PyData CoBo: Kickoff

Hello Pythonistas!

We are happy to welcome you to the kickoff of a new PyData chapter in Cologne and Bonn. Starting right off in the new year, there will be monthly Meetups in both cities alternately with room for networking, free refreshments and of course many exciting talks around the topic of Python programming in the field of data science.

The first event is hosted by inovex GmbH in Cologne, where we will hear two talks from experts in different domains.

The first talk will be given by Alexander Hendorf from Königsweg. (https://www.koenigsweg.com/koenigsweg/berater/ahendorf/index_ger.html). Alex is an expert in Big Data, Data Mining and Machine Learning. As a program Chair in EuroPython, PyConDe, PyData Karlsruhe and more he is an experienced and well-known core member of the PyData Community.

The second talk is given by Dr. Florian Wilhelm, Principal Data Scientist at inovex (https://www.inovex.de/blog/author/fwilhelm/). He works with Python tools in the field of Predictive & Prescriptive Analysis and Big Data. He will give a talk about Uncertainty quantification in deep learning.

Talk #1: Alexander Hendorf - Natural Language Processing for Managers

Artificial intelligence as Natural Language Processing promises great value. The technology is mostly understood only by few, yet still unexplainable even for experts.
In this talk I'll present how to narrow the hype down to real value explainable to everyone in your organisation - without the math.
In this talk I'll present a use case how KÖNIGSWEG made a large corpus of documents stored in our clients drives accessible and experienceable with Natural Language Processing. Our client has a large collection of research-related documents (Word, PPTX, PDFs, ePubs, etc.) stored on the company's network drives. Our client wanted a better understanding of the contents collected over decades and especially a deeper understanding of relationships between documents and topics. We built a solution featuring Spacy, Gensim, some Neural Nets and an UI to provide a better and intuitive access to the documents.

Talk #2: Dr. Florian Wilhelm - Uncertainty quantification in deep learning

With the advent of Deep Learning (DL), the field of AI made a giant leap forward and it is nowadays applied in many industrial use-cases. Especially critical systems like autonomous driving, require that DL methods not only produce a prediction but also state the certainty about the prediction in order to assess risks and failure.
In my talk, I will give an introduction to different kinds of uncertainty, i.e. epistemic and aleatoric. To have a baseline for comparison, the classical method of Gaussian Processes for regression problems is presented. I then elaborate on different DL methods for uncertainty quantification like Quantile Regression, Monte-Carlo Dropout, and Deep Ensembles. The talk is concluded with a comparison of these techniques to Gaussian Processes and the current state of the art.

Agenda
18:30 - 19:00: Doors open
19:00 - 19:10: Introduction
19:10 - 19:40: Talk #1
19:40 - 20:30: Pizza + Networking
20:30 - 21:00: Talk #2

Place:
inovex GmbH
Schanzenstraße 6-20, Kupferhütte 1.13, 51063 Köln