Seminars
The DIG team holds a seminar about every two weeks with speakers either from the team, or invited.
You can add the seminars to your calendar with this ics file, and get emails about future seminars by subscribing to our mailing-list.
If you would like to present your work at our seminar, please contact Nils.
Upcoming Seminars
CIPHER: Towards Sovereign Blockchain Intelligence for Europe
Tuesday, November 17, 2026 11:45, 4A301
Peter Fratric (European Commission’s Joint Research Centre)
In an era of heightened geopolitical uncertainty, the European Union is pursuing greater strategic autonomy in critical digital technologies. As crypto-assets and DeFi continue to reshape global markets, developing sovereign analytical capabilities is increasingly important for European financial stability, supervisory independence and monetary sovereignty. Initiated by the European Commission’s Joint Research Centre, CIPHER (Crypto Intelligence Platform for a Holistic European Regulation) provides a scalable intelligence architecture designed to translate complex on-chain data into actionable policy insights. By ingesting data directly from blockchain nodes and processing it through dedicated storage, indexing and analytics layers, CIPHER provides the foundation for near-real-time monitoring of crypto-asset markets, transfers and protocol activity. Current development focuses on advancing in-house methodologies for entity resolution and attribution, extending transfer-level monitoring capabilities, and addressing technical challenges related to blockchain interoperability and cross-chain analysis. In parallel, CIPHER is laying the groundwork for longer-term risk-assessment and stress-testing capabilities, while exploring greater data interoperability across EU institutions and strengthening collaboration with academia. The presentation provides an update on these developments and demonstrates how CIPHER is evolving into a sovereign European capability for crypto-market intelligence and evidence-based regulatory decision-making.
Riding the Waves of Change: Towards Streaming Continual Learning
Wednesday, November 18, 2026 11:45, 4A301
Federico Giannini (Politecnico di Milano)
Real-world data streams continuously evolve, requiring Machine Learning models to adapt to changing distributions while preserving previously acquired knowledge. Some streaming scenarios, such as sensor data streams, also exhibit temporal dependencies. This setting is defined as Evolving Streaming Time Series (ESTS).
Scalable Graph Edit Distance with Deep Reinforcement Learning
Tuesday, December 01, 2026 11:45, 4A301
Hector Kohler
In this talk, we will tackle the Graph Edit Distance (GED) problem. Given two labeled graphs, how many edits does one need to get from one graph to the other (e.g. remove edge, add node, relabel nodes…)? This is an NP-hard problem for which exact solvers can’t scale past graphs with ~20 nodes. In recent years, graph machine learning has shown good performances in the supervised learning setting for the GED. GNN models are trained on graph edit distance instances with up to ~20 nodes, for which exact labels can be computed, and then are used at test time to solve larger and larger instances. Unsuprisingly, performances decay with problem size. In parallel, optimal transport approaches were also studied. They formulate GED as a graph matching problem that can be approximated in cubic time and quadratic space and where supervised training labels are not necessary. The OT quadratic space complexity prohibits scaling to 1000s nodes. We propose to use deep reinforcement learning to train policies that can approximate GED in quadratic time and linear space. We show empirically that our approach can compute GED at scale better than all other existing work.
Title TBA
Wednesday, December 09, 2026 11:45, 4A301
Steffen Staab
Abstract TBA
Past Seminars
From Observations to Explanations: Exploring Methodologies for Identifying Relevant Actual Causes in Complex Systems
Tuesday, October 27, 2026 11:45, 4A301
Samuel Reyd
Complex systems such as multi-agent and cyber-physical systems exhibit emergent behaviors that are hard to understand and explain. Causal models offer a principled framework for this, but building one for a complex system is already difficult: the relevant scope, abstraction scale, and variables are not given in advance. Moreover, such models capture general causation, i.e., which factors influence a class of outcomes, whereas a user asking “why did this happen?” needs an actual cause: the specific facts responsible for a specific event. Actual causation is conceptually and computationally demanding; finding actual causes is intractable in general, and not all causes are equally informative. Producing a useful explanation therefore requires three steps: modeling the system causally, identifying its actual causes, and filtering them for relevance. This model-identify-filter pipeline forms the backbone of the thesis. This thesis first reviews the challenges of causal modeling in complex adaptive systems and examines the choice of abstraction scale through causal emergence, showing that emergence detection depends heavily on the sampling distribution used. It then addresses the identification of actual causes, proposing a data-based method for Markovian multivariate systems and approximate search algorithms with adjustable precision that recover the full set of actual causes in general SCMs, released as an open-source Python module. Finally, it presents a general, tunable method for selecting relevant causes according to user and context, and develops a normality criterion to produce more relevant explanations. Overall, the thesis connects formal causal models, scalable computation, and human-centred explanation into a coherent framework for context-aware explanation in complex systems.
EMNLP+ISWC talks
Tuesday, October 20, 2026 11:45, 4A301
DIG team
We will hear about the papers accepted at EMNLP and ISWC 2026.
Toward Adaptive Intelligence
Tuesday, October 13, 2026 11:45, 4A301
Nilesh Verma (University of Waikato)
Real-world data arrives as an endless stream whose distribution shifts over time. Models that work today may degrade tomorrow, and the human effort required to repeatedly re-tune them does not scale. This talk presents a line of work on adaptive intelligence, where systems select, configure, and revise their own learning pipelines as the data changes. I begin with AutoML for non-stationary data streams, covering online pipeline search, automated drift and outlier handling, and meta-learning for data streams. I then turn to tabular foundation models and ask what happens when in-context learners encounter concept drift. Along the way, I introduce TuiML, an open-source MCP-native ML runtime that enables agents and researchers to run, benchmark, and empirically study machine learning workflows.
AI, Proofs, and Programs
Tuesday, September 08, 2026 11:45, 4A301
Thomas Bonald
AI is transforming mathematics, raising profound questions: Can we trust AI-generated proofs? Do we understand them? And what role will mathematicians play in this new era? This talk explores these questions through the lens of proof checking, formal systems, and the elegant Curry-Howard correspondence—a deep connection between logic, computation, and the nature of mathematical truth.
Identifying and Evaluating Misleading Climate Communication with Natural Language Processing
Tuesday, June 23, 2026 14:00, 4A301
Tom Calamai
Climate communication is becoming more abundant, but not necessarily more informative. This thesis investigates whether Natural Language Processing (NLP) can help structure climate-related discourse, distinguish substantive content from vague or rhetorical formulations, and support credibility assessment. By examining the literature on greenwashing and major datasets for climate-related NLP tasks, it highlights key limitations, including subjectivity, ambiguity, and noisy data. It then proposes ways to address these issues through annotation schemes and evaluation metrics designed for ambiguity, as well as methods for propagating uncertainty into downstream analyses. Overall, the thesis shows that NLP can make climate-related discourse more explicit and analyzable, while also emphasizing that progress depends not only on model performance, but also on task design, data quality, and uncertainty-aware evaluation.
Uncertainty in the Data Engineering Pipeline: From Imputation to Explanation
Tuesday, June 02, 2026 11:45, 4A301
Julia Stoyanovich (New York University)
Machine learning pipelines are commonly evaluated on model accuracy, yet much of the uncertainty that shapes real-world outcomes originates not in the model itself but in the data engineering decisions that precede it. This talk examines two manifestations of that upstream uncertainty and their downstream consequences for trustworthy AI.