Program Overview

Programme Overview

Day 1 (Apr 22, 2026)

09:30–10:30 Keynote Geoff Webb: “Convolutional kernels for effective and scalable time series analytics “
10:30–11:00 Break Coffee break
11:00–12:00 Session Plenary session — Neural networks (3 talks)
  1. T-SE: A method built on Squeeze-and-Excitation mechanisms for Convolutional Neural Networks’ energy efficiency
  2. PromptFusionSR: Multimodal Enhancement of Low-Resolution Images with Automatic Prompt-Guided Diffusion
  3. Guided Feature Distillation for YOLO11: Efficient Detection of Pseudomonas aeruginosa in High-Resolution Microscopy
12:00–13:30 Lunch Lunch
13:30–15:00 Session Plenary session — Time series and temporal data (4 talks)
  1. An End-to-End Framework for Measuring Product Cannibalization using Multivariate Time Series Forecasting
  2. The Window Dilemma: Why Concept Drift Detection is Ill-Posed
  3. Toward Improved Time-Series Explanations for Federated Learning in Healthcare
  4. Grasynda: Graph-based Synthetic Time Series Generation
15:00–15:30 Break Coffee break
15:30–16:30 Keynote Isabel Valera: “Causal Generative Models: From theory to practice”
16:30–17:00 Move Move location
17:00–19:00 Event Welcome reception (city hall) and Introduction of mentors to PhD Forum Participants

Day 2 (Apr 23, 2026)

09:00–10:00 Keynote Ana Lucic: “A foundation model of the Earth system”
10:00–11:00 Session Plenary session — Graphs (3 talks)
  1. Learning Molecular Structures from Infrared Spectra through Latent Evidence Prediction
  2. Conditional Motif-based Graph Convolutional Network for Anomaly Detection in the Waste Management Network
  3. Graph Neural Networks for Graph-Level Regression on Heterogeneous Network Data: Use Case in Early-Stage Optimization of Software Mapping on Multicore Platforms
11:00–11:30 Poster Pitches & Move Poster Pitches (1 minute) and Move location
11:30–13:30 Event Lunch buffet + poster session
13:30–15:00 Session Plenary session — Applications (4 talks)
  1. Predicting and Interpolating Spatiotemporal Environmental Data: A Case Study of Groundwater Storage in Bangladesh
  2. Combining Dynamic Bayesian Networks with Population Dynamics Modelling to Predict Breeding Success in Seabirds
  3. Bridging Forecast Accuracy and Inventory KPIs: A Simulation-Based Evaluation Framework
  4. LeTMEMo: Leveraging Topic Modeling for Enhancing (Closed-Vocabulary) Models
15:00–15:20 Break Coffee break
15:20–17:00 Session Plenary session — Decision trees & data (5 talks)
  1. Deep Decision Forest
  2. A Generic Complete Anytime Beam Search for Optimal Decision Tree
  3. CARTGen-IR: Synthetic Tabular Data Generation for Imbalanced Regression
  4. Enabling Context-Aware Data Reduction
  5. Drop the mask! GAMM – A Taxonomy for Graph Attributes Missing Mechanisms
17:00–18:00 Event Beach BBQ

Day 3 (Apr 24, 2026)

09:00–10:00 Keynote Fredrik Heintz: “Towards Trustworthy and Factual Large Language Models”
10:00–10:40 Session Plenary session — Patterns (2 talks)
  1. Heterogeneous Pattern Sampling according to Frequency
  2. Exceptional Model Residual Mining, and Three Richer EMM Description Languages
10:40–11:10 Break Coffee break
11:10–12:10 Session Plenary session — Clustering & community detection (3 talks)
  1. E-PAGEC: A Differentiable Joint Attributed-Graph Embedding and Clustering Model
  2. Adaptive Local Kernel for Efficient Active Pairwise Constraint Clustering
  3. Fair multilayer community detection: A research agenda
12:10–13:30 Lunch Lunch
13:30–14:50 Session Plenary session — Foundations (4 talks)
  1. Evidential Deep Learning is not Evidential Learning: A Clear Distinction
  2. Detecting Propensity Score Shifts Across Groups in Positive–Unlabeled Data
  3. On Sample-Wise Strict Monotonicity with a Gradient Update
  4. Exploiting Treatment Similarities for Enhanced Multi-Treatment Uplift Prediction
14:50–15:20 Break Coffee break
15:20–17:00 Session Plenary session — XAI (5 talks)
  1. NTS-DAGMA: A Score-Based Causal Discovery for Anomaly Detection
  2. Analyzing Shapley Additive Explanations to Understand Anomaly Detection Algorithms Behaviors and Their Complementarity
  3. Counterfactual Explanations for Conformal Regression Intervals
  4. Extending Information Bottleneck Attribution to Video Sequences for Deepfake Detection
  5. Fast Model Selection for Interpretable Gaussian Process Models Using Laplace Approximation
17:00–18:00 Event Good bye drinks (informal)