Contacts (if not stated otherwise): Markus Schedl.

Remark: We are open for new proposals - if you are interested in Recommender Systems, feel free to contact us!

Topics:

• Extracting music listening intents/purposes from music-related and behavioral data for user modeling
• Fairness of recommendation algorithms
• Disentangled representation learning for recommendation
• Removing unwanted bias/information in deep neural networks using information theory
•  Debiasing deep models using Siamese neural networks
• Debiasing graph-based models (graph neural networks)
• Multistakeholder recommender systems
• Multi-objective bias mitigation in recommender systems
• Popularity bias in recommender systems
• Explainability in recommender systems
• Robustness of recommendation algorithms
• Differential privacy in recommendation algorithms
• Carbon footprint of recommender system algorithms
• Psychology-informed recommender systems (cognition models, affect-aware, personality-aware systems)
• Psychology-inspired sampling strategies for collaborative filtering
• Post-processing collaborative filtering results according to psychological curiosity-arousal-model
• Session-based recommendation
• Sequential recommendation
• Recommender systems based on large language models
• Recommender systems based on diffusion models
• Adversarial training for unlearning protected user characteristics in DNN-based RecSys
• Autoencoders for recommender systems
• Simulating the long-term impact of recommendation algorithms
• Conversational recommender systems
• Domain-specific recommender systems, e.g., music, jobs, point-of-interest, accommodation, fashion, etc.
• User studies on bias and fairness of recommender systems