12 papers
Transfer Learning for Evolving Domains
Ricardo Ribeiro Pereira, Jacopo Bono, Hugo Ferreira +4
Transfer learning explores how to leverage knowledge from various tasks or domains (sources) to enhance predictive performance in related tasks or domains (targets). Typically, tra…
Causal Discovery on Irregular Time Series
Martim Penim, Ricardo Ribeiro Pereira, Jacopo Bono +3
Causal discovery methods have shown strong performance in temporal systems, but they typically rely on regular and discrete lag structures, limiting their applicability to regularl…
Decoupling Inference from State Updates in Low-Latency Feature Engines via Probabilistic Thinning
Augusto Peres, Iker Perez, Pedro Valdeira +4
Streaming data systems increasingly underpin Machine Learning workflows that maintain large numbers of continuously updated aggregations. In production settings, each incoming even…
Rethinking XAI Evaluation: A Human-Centered Audit of Shapley Benchmarks in High-Stakes Settings
Inês Oliveira e Silva, Sérgio Jesus, Iker Perez +4
Shapley values are a cornerstone of explainable AI, yet their proliferation into competing formulations has created a fragmented landscape with little consensus on practical deploy…
MUSE: Multi-Tenant Model Serving With Seamless Model Updates
Cláudio Correia, Alberto E. A. Ferreira, Lucas Martins +7
In binary classification systems, decision thresholds translate model scores into actions. Choosing suitable thresholds relies on the specific distribution of the underlying model…
Evaluating Transfer Learning Methods on Real-World Data Streams: A Case Study in Financial Fraud Detection
Ricardo Ribeiro Pereira, Jacopo Bono, Hugo Ferreira +3
When the available data for a target domain is limited, transfer learning (TL) methods can be used to develop models on related data-rich domains, before deploying them on the targ…
Mind the truncation gap: challenges of learning on dynamic graphs with recurrent architectures
João Bravo, Jacopo Bono, Pedro Saleiro +2
Systems characterized by evolving interactions, prevalent in social, financial, and biological domains, are effectively modeled as continuous-time dynamic graphs (CTDGs). To manage…
Fair-OBNC: Correcting Label Noise for Fairer Datasets
Inês Oliveira e Silva, Sérgio Jesus, Hugo Ferreira +4
Data used by automated decision-making systems, such as Machine Learning models, often reflects discriminatory behavior that occurred in the past. These biases in the training data…
Deep-Graph-Sprints: Accelerated Representation Learning in Continuous-Time Dynamic Graphs
Ahmad Naser Eddin, Jacopo Bono, David Aparício +3
Continuous-time dynamic graphs (CTDGs) are essential for modeling interconnected, evolving systems. Traditional methods for extracting knowledge from these graphs often depend on f…
Adversarial training for tabular data with attack propagation
Tiago Leon Melo, João Bravo, Marco O. P. Sampaio +4
Adversarial attacks are a major concern in security-centered applications, where malicious actors continuously try to mislead Machine Learning (ML) models into wrongly classifying…
From random-walks to graph-sprints: a low-latency node embedding framework on continuous-time dynamic graphs
Ahmad Naser Eddin, Jacopo Bono, David Aparício +4
Many real-world datasets have an underlying dynamic graph structure, where entities and their interactions evolve over time. Machine learning models should consider these dynamics…
Lightweight Automated Feature Monitoring for Data Streams
João Conde, Ricardo Moreira, João Torres +5
Monitoring the behavior of automated real-time stream processing systems has become one of the most relevant problems in real world applications. Such systems have grown in complex…