PyData Heidelberg #12: Convex Optimization with DSP & xAI with SHAP + Lime

DataScience and AI: in person in Heidelberg and live on PyData.TV on YouTube

Agenda
18:00 Doors open
18:30 Welcome
18:45 Philipp Schiele - Introducing Disciplined Saddle Programming (DSP): A New Paradigm in Convex Optimization
19:15 Break: Networking with snacks and beverages
20:00 Christophe Krech - Unveiling the Black Box: Exploring Explainable AI with SHAP and Lime for Tabular Data in Python
20:30 Lightning Talks
20:45 Networking with snacks and beverages
21:30 End

Lightning Talks
Join us by contributing a five-minute lightning talk!
Fill out this form.

How to sign up for on site
It's important for us to make this meet up happen in a responsible way. We have limited seats available only.
No limits to sign up remotely!

How to join remotely
Join the live stream on YouTube.

Q&A
Ask via Slido
This event will be in English.
----
Talk #1
Philipp Schiele (Ludwig Maximilian University of Munich)
Introducing Disciplined Saddle Programming (DSP): A New Paradigm in Convex Optimization
Disciplined Saddle Programming (DSP), a new Python-based domain-specific language, significantly enhances the approach to convex-concave saddle problems, crucial in fields like game theory, machine learning, and finance. Hosted on GitHub, DSP extends the CVXPY framework, streamlining the dualization process in optimization.
DSP focuses on robust optimization problems, providing an intuitive interface for problem specification and resolution. It builds upon the conic-representable saddle programs by Juditsky and Nemirovski, applying disciplined convex programming to saddle problems.
DSP's introduction is a call to the broader scientific and engineering communities to explore its diverse applications. It simplifies complex optimization tasks, making them more accessible and manageable, and holds the potential to significantly impact various optimization-reliant fields.

Philipp Schiele's educational background is in finance and economics and he is currently pursuing a PhD in financial econometrics at the Ludwig Maximilian University of Munich, where he taught various courses in statistics. He is a CVXPY maintainer and has presented a tutorial at SciPy 2022. Generally, he is enthusiastic about finance, optimization, and technology, especially open-source projects.
Apart from that, he also conducts workshops at SciPy US on "Controlling Self-Landing Rockets Using CVXPY" 🚀

Talk #2
Christophe Krech - Unveiling the Black Box: Exploring Explainable AI with SHAP and Lime for Tabular Data in Python
In the era of complex machine learning models, understanding and interpreting their decisions is crucial for fostering trust and transparency; and will become a regulatory requirement for many applications with the implementation of the EU AI act. Explainable AI (XAI) specifically tailored for tabular data can demystify the black box of machine learning. SHAP (SHapley Additive exPlanations) and Lime (Local Interpretable Model-agnostic Explanations) are to very powerful model-agnostic tools to enhance the understanding of model predictions, troubleshoot biases, and communicate machine learning insights effectively. Thanks to great open-source implementations, they can also be seamlessly integrated into existing Python workflows in many real-world applications.

Christophe Krech is a senior data scientist at Experian. During his studies in Mannheim and Darmstadt, he already focused on the explainability of machine learning methods and the associated regulatory challenges. After completing his master’s degree in data science, he joined the global information service provider Experian in 2019. Since then, he has been supporting FinTechs, e-commerce retailers and banks in the successful use of machine learning for risk management. Explainability of machine learning models and their implementation in Python are an integral part of his work there.
----
Lightning Talks:
1. Your spot! Submit a talk here
2. Your spot! Submit a talk here
3. Your spot! Submit a talk here

Acknowledgements
Also a big thank you to our sponsors:

• HEIDELBERG MATERIALS, for hosting the meetup.
• KÖNIGSWEG, for supporting the organization.
• NUMFOCUS, for promoting open source software.

Contact
If you have any questions or suggestions, please feel free to contact us via:

• Meetup
• Want to speak? Submit a talk here.
• Interested in hosting an event? Here's our Info-Deck & contact to the organisers!