Zobrazeno 1 - 10
of 59
pro vyhledávání: '"P Quinzan"'
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 variable (IV) regression, which utilises a key unconfounded variable kn
Externí odkaz:
http://arxiv.org/abs/2405.08498
Learning the causes of time-series data is a fundamental task in many applications, spanning from finance to earth sciences or bio-medical applications. Common approaches for this task are based on vector auto-regression, and they do not take into ac
Externí odkaz:
http://arxiv.org/abs/2311.06012
Autor:
Mamaghan, Amir Mohammad Karimi, Dittadi, Andrea, Bauer, Stefan, Johansson, Karl Henrik, Quinzan, Francesco
Causal reasoning can be considered a cornerstone of intelligent systems. Having access to an underlying causal graph comes with the promise of cause-effect estimation and the identification of efficient and safe interventions. However, learning causa
Externí odkaz:
http://arxiv.org/abs/2311.05421
Knowing the features of a complex system that are highly relevant to a particular target variable is of fundamental interest in many areas of science. Existing approaches are often limited to linear settings, sometimes lack guarantees, and in most ca
Externí odkaz:
http://arxiv.org/abs/2306.07024
The contribution of this paper is a generalized formulation of correctional learning using optimal transport, which is about how to optimally transport one mass distribution to another. Correctional learning is a framework developed to enhance the ac
Externí odkaz:
http://arxiv.org/abs/2304.01701
Notions of counterfactual invariance (CI) have proven essential for predictors that are fair, robust, and generalizable in the real world. We propose graphical criteria that yield a sufficient condition for a predictor to be counterfactually invarian
Externí odkaz:
http://arxiv.org/abs/2207.09768
Autor:
Quinzan, Francesco, Khanna, Rajiv, Hershcovitch, Moshik, Cohen, Sarel, Waddington, Daniel G., Friedrich, Tobias, Mahoney, Michael W.
We study the fundamental problem of selecting optimal features for model construction. This problem is computationally challenging on large datasets, even with the use of greedy algorithm variants. To address this challenge, we extend the adaptive qu
Externí odkaz:
http://arxiv.org/abs/2202.13718
Several large-scale machine learning tasks, such as data summarization, can be approached by maximizing functions that satisfy submodularity. These optimization problems often involve complex side constraints, imposed by the underlying application. I
Externí odkaz:
http://arxiv.org/abs/2102.06486
Autor:
Amir Mohammad Karimi Mamaghan, Andrea Dittadi, Stefan Bauer, Karl Henrik Johansson, Francesco Quinzan
Publikováno v:
Entropy, Vol 26, Iss 7, p 556 (2024)
Causal reasoning can be considered a cornerstone of intelligent systems. Having access to an underlying causal graph comes with the promise of cause–effect estimation and the identification of efficient and safe interventions. However, learning cau
Externí odkaz:
https://doaj.org/article/96c8c35d649b4a76978408c26470f586
Autor:
Doskoč, Vanja, Friedrich, Tobias, Göbel, Andreas, Neumann, Frank, Neumann, Aneta, Quinzan, Francesco
We study the problem of maximizing a non-monotone submodular function under multiple knapsack constraints. We propose a simple discrete greedy algorithm to approach this problem, and prove that it yields strong approximation guarantees for functions
Externí odkaz:
http://arxiv.org/abs/1911.06791