6 papers · 1 filter
MIBP-Cert: Certified Training against Data Perturbations with Mixed-Integer Bilinear Programs
Tobias Lorenz, Marta Kwiatkowska, Mario Fritz
Data errors, corruptions, and poisoning attacks during training pose a major threat to the reliability of modern AI systems. While extensive effort has gone into empirical mitigati…
Learning Decision Policies with Instrumental Variables through Double Machine Learning
Daqian Shao, Ashkan Soleymani, Francesco Quinzan +1
A common issue in learning decision-making policies in data-rich settings is spurious correlations in the offline dataset, which can be caused by hidden confounders. Instrumental v…
FullCert: Deterministic End-to-End Certification for Training and Inference of Neural Networks
Tobias Lorenz, Marta Kwiatkowska, Mario Fritz
Modern machine learning models are sensitive to the manipulation of both the training data (poisoning attacks) and inference data (adversarial examples). Recognizing this issue, th…
The Trembling-Hand Problem for LTLf Planning
Pian Yu, Shufang Zhu, Giuseppe De Giacomo +2
Consider an agent acting to achieve its temporal goal, but with a "trembling hand". In this case, the agent may mistakenly instruct, with a certain (typically small) probability, a…
Uncertainty-Aware Explanations Through Probabilistic Self-Explainable Neural Networks
Jon Vadillo, Roberto Santana, Jose A. Lozano +1
The lack of transparency of Deep Neural Networks continues to be a limitation that severely undermines their reliability and usage in high-stakes applications. Promising approaches…
STR-Cert: Robustness Certification for Deep Text Recognition on Deep Learning Pipelines and Vision Transformers
Daqian Shao, Lukas Fesser, Marta Kwiatkowska
Robustness certification, which aims to formally certify the predictions of neural networks against adversarial inputs, has become an integral part of important tool for safety-cri…