Explainable AI - Training Week
The course provides a comprehensive introduction to Explainable Artificial Intelligence (XAI), emphasizing the methodologies and practical applications of cutting-edge models such as LIME, SHAP, deep learning XAI, time series-based XAI methods and others.
Time & Location
06 Oct 2025, 08:30 – 10 Oct 2025, 18:00
BFH Business School, Brückenstrasse 73, 3005 Bern, Switzerland
About the event
The training week provides a comprehensive introduction to Explainable Artificial Intelligence (XAI), emphasizing both foundational and advanced methodologies. Participants will explore practical applications of model-agnostic techniques such as Partial Dependence Plots (PDP), Individual Conditional Expectation (ICE) plots, LIME, and SHAP, alongside specialized approaches for interpreting complex deep learning models. Particular attention will be given to explainability in neural networks, including gradient-based methods, as well as emerging XAI techniques tailored to time-series data. Participants will explore how these techniques enhance interpretability and transparency in AI systems, along with the challenges they face, such as scalability, interpretability trade-offs, and accuracy limitations.
The course also investigates the limitations and reliability of XAI models when applied to complex datasets, with advanced discussions on their performance and practical constraints. A distinctive focus is placed on financial applications, examining how XAI can address the unique challenges and regulatory requirements of the financial sector.
By the end of the course, participants will gain the expertise to implement XAI models, critically evaluate their effectiveness, and apply them responsibly within financial systems, fostering trust and compliance with regulatory standards.
Modules and Credits: The training week covers the following models and credits of the MSCA doctoral program:
6-9th October, 2025: Foundation Module "Need for Explainable AI in Finance" (4 ECT)
10th October, 2025: Module "Research Ethics and Sustainable Research Management" (1 ECT)
Prior Knowledge: Participants are expected to have a foundational understanding of machine learning concepts, including supervised and unsupervised learning, common algorithms, and evaluation metrics. Certain familiarity with Python programming is needed, with experience using libraries such as scikit-learn, pandas, and NumPy. Familiarity with basic statistics and linear algebra will also be helpful for understanding the mathematical foundations of explainability methods.
Detailed Schedule
Monday, 6th October
8:30 - 9:00: Networking Coffee
9:00 - 9:15: Welcome to BFH (Christian Hopp, Head of Research, Bern University of Applied Sciences)
9:15 - 9:45: Overview of the Week / Launch of Training Competition (Branka Hadji Misheva & Lucia Gomez)
Break
Industry Session - AI in Action (real-world use cases)
10:00 - 10:30: Explainable AI for Philanthropy and Finance (Milos Maricic, CEO, Altruist League)
10:30 - 11:00: Representative from Microsoft Switzerland
11:00 - 11:30: Alexandru-Septimiu Rif, Portfolio Manager, Alean Capital AG and Robert Gutsche, Professor Applied Data Science, Bern University of Applied Sciences
11:30 - 12:00: Lilian Nordet, Director of HUB+
Lunch
13:30 - 14:00: Aidas Malakauskas, Head of Omnichannel Solutions, Swedbank
14:00 - 14:30: Fabio Duo, Founder & CEO, PeakPrivacy and Founder & Project Manager, Freihandlabor GmbH
14:30 - 15:00: Luba Schoenig, Co-Founder, Umushroom and former banker, Credit Suisse
Break
15:30 - 16:00: Stefan Theussl, Head of Research Innovation Hub, Raiffeisen Bank International AG
16:00 - 16:30: Gennaro Di Brino, Head of Data Science, Cardo AI
Break
17:00 - 18:00: Student Pitches xAI
Tuesday, 7th October
Wednesday, 8th October
Thursday, 9th October
Friday, 10th October
Speakers and Talks
Milos Maricic (CEO, Altruist League): Explainable AI for Philanthropy and Finance
Stefan Theussl (Head of Research Innovation Hub; Raiffeisen Bank International AG): (Explainable) AI in Practice: Real-World Applications at RBI
Luba Schoenig (Formar Credit Suiss banker & co-founder of Umushroom): AI in Finance: Redefining Intelligence in Investing
Alexandru-Septimiu Rif (Portfolio Manager at Alean Capital AG)
Robert Gutsche (Professor in Applied Data Science and Finance at BFH)
Gennaro Di Brino (Head of Data Science at CARDO AI)
Fabio Duò (Founder of PeakPrivacy)
Aidas Malakauskas (Head of Omnichannel Solutions, Swedbank): Applied AI in Banking: Failures, Successes, and What’s Next?
Branka Hadji Misheva (Professor in Applied Data Science at BFH): SHAP and LIME: Deep Dive
Lucia Teijeiro Gomez (Professor in Applied Data Science at BFH): SHAP and LIME: Deep Dive
Julius Kooistra (Researcher in Applied Data Science at BFH): XAI in Public Employment Services
Faizan Ahmed (Lecturer and program director in Business Information Technology at the University of Twente): Explainable AI for Deep Learning & Time Series
Golnoosh Babaei (Senior researcher at University of Pavia): SAFE AI
Accommodation Options
To help with planning your stay, we have compiled a list of affordable accommodation options in Bern, which are near the university campus.