Zobrazeno 1 - 10
of 1 464
pro vyhledávání: '"van der Schaar Mihaela"'
We consider the problem of quantifying how an input perturbation impacts the outputs of large language models (LLMs), a fundamental task for model reliability and post-hoc interpretability. A key obstacle in this domain is disentangling the meaningfu
Externí odkaz:
http://arxiv.org/abs/2412.00868
Autor:
Seedat, Nabeel, Tozzi, Caterina, Ardiaca, Andrea Hita, van der Schaar, Mihaela, Weatherall, James, Taylor, Adam
The reuse of historical clinical trial data has significant potential to accelerate medical research and drug development. However, interoperability challenges, particularly with missing medical codes, hinders effective data integration across studie
Externí odkaz:
http://arxiv.org/abs/2411.13163
Publikováno v:
Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, Canada, 2024
Estimating causal quantities from observational data is crucial for understanding the safety and effectiveness of medical treatments. However, to make reliable inferences, medical practitioners require not only estimating averaged causal quantities,
Externí odkaz:
http://arxiv.org/abs/2411.03387
Mathematical optimization is fundamental to decision-making across diverse domains, from operations research to healthcare. Yet, translating real-world problems into optimization models remains a formidable challenge, often demanding specialized expe
Externí odkaz:
http://arxiv.org/abs/2411.01679
Deep learning sometimes appears to work in unexpected ways. In pursuit of a deeper understanding of its surprising behaviors, we investigate the utility of a simple yet accurate model of a trained neural network consisting of a sequence of first-orde
Externí odkaz:
http://arxiv.org/abs/2411.00247
Real-world machine learning systems often encounter model performance degradation due to distributional shifts in the underlying data generating process (DGP). Existing approaches to addressing shifts, such as concept drift adaptation, are limited by
Externí odkaz:
http://arxiv.org/abs/2411.00186
Schema matching -- the task of finding matches between attributes across disparate data sources with different tables and hierarchies -- is critical for creating interoperable machine learning (ML)-ready data. Addressing this fundamental data-centric
Externí odkaz:
http://arxiv.org/abs/2410.24105
The predominant de facto paradigm of testing ML models relies on either using only held-out data to compute aggregate evaluation metrics or by assessing the performance on different subgroups. However, such data-only testing methods operate under the
Externí odkaz:
http://arxiv.org/abs/2410.24005
Digital Twins (DTs) are computational models that simulate the states and temporal dynamics of real-world systems, playing a crucial role in prediction, understanding, and decision-making across diverse domains. However, existing approaches to DTs of
Externí odkaz:
http://arxiv.org/abs/2410.23691
Autor:
Feuerriegel, Stefan, Frauen, Dennis, Melnychuk, Valentyn, Schweisthal, Jonas, Hess, Konstantin, Curth, Alicia, Bauer, Stefan, Kilbertus, Niki, Kohane, Isaac S., van der Schaar, Mihaela
Publikováno v:
Nature Medicine, vol. 30, pp. 958-968 (2024)
Causal machine learning (ML) offers flexible, data-driven methods for predicting treatment outcomes including efficacy and toxicity, thereby supporting the assessment and safety of drugs. A key benefit of causal ML is that it allows for estimating in
Externí odkaz:
http://arxiv.org/abs/2410.08770