1 code implementation • 15 Apr 2024 • Margherita Lazzaretto, Jonas Peters, Niklas Pfister
We consider the task of predicting a response Y from a set of covariates X in settings where the conditional distribution of Y given X changes over time.
1 code implementation • 15 Feb 2024 • Niklas Pfister, Peter Bühlmann
In this work, we extend the nonparametric statistical model to explicitly allow for extrapolation and introduce a class of extrapolation assumptions that can be combined with existing inference techniques to draw extrapolation-aware conclusions.
1 code implementation • 30 Nov 2023 • Anton Rask Lundborg, Niklas Pfister
Existing statistical methods for compositional data analysis are inadequate for many modern applications for two reasons.
no code implementations • 9 Oct 2023 • Nicola Gnecco, Jonas Peters, Sebastian Engelke, Niklas Pfister
In particular, we establish a novel connection between the field of distribution generalization from machine learning, and simultaneous equation models and control function from econometrics.
no code implementations • 6 Oct 2023 • Sorawit Saengkyongam, Elan Rosenfeld, Pradeep Ravikumar, Niklas Pfister, Jonas Peters
In this paper, we consider the task of intervention extrapolation: predicting how interventions affect an outcome, even when those interventions are not observed at training time, and show that identifiable representations can provide an effective solution to this task even if the interventions affect the outcome non-linearly.
no code implementations • 19 Jun 2023 • Sorawit Saengkyongam, Niklas Pfister, Predrag Klasnja, Susan Murphy, Jonas Peters
A major challenge in policy learning is how to adapt efficiently to unseen environments or tasks.
1 code implementation • 15 May 2022 • Shimeng Huang, Elisabeth Ailer, Niki Kilbertus, Niklas Pfister
We propose KernelBiome, a kernel-based nonparametric regression and classification framework for compositional data.
no code implementations • 17 Mar 2022 • Niklas Pfister, Jonas Peters
Exogenous heterogeneity, for example, in the form of instrumental variables can help us learn a system's underlying causal structure and predict the outcome of unseen intervention experiments.
1 code implementation • 12 Feb 2022 • Sebastian Weichwald, Søren Wengel Mogensen, Tabitha Edith Lee, Dominik Baumann, Oliver Kroemer, Isabelle Guyon, Sebastian Trimpe, Jonas Peters, Niklas Pfister
Questions in causality, control, and reinforcement learning go beyond the classical machine learning task of prediction under i. i. d.
Model-based Reinforcement Learning reinforcement-learning +1
1 code implementation • 3 Feb 2022 • Sorawit Saengkyongam, Leonard Henckel, Niklas Pfister, Jonas Peters
Most of the existing estimators assume that the error term in the response $Y$ and the hidden confounders are uncorrelated with the instruments $Z$.
1 code implementation • 1 Jun 2021 • Sorawit Saengkyongam, Nikolaj Thams, Jonas Peters, Niklas Pfister
We adopt the concept of invariance from the causality literature and introduce the notion of policy invariance.
1 code implementation • 12 Jun 2020 • Rune Christiansen, Niklas Pfister, Martin Emil Jakobsen, Nicola Gnecco, Jonas Peters
We introduce the formal framework of distribution generalization that allows us to analyze the above problem in partially observed nonlinear models for both direct interventions on $X$ and interventions that occur indirectly via exogenous variables $A$.
Methodology Primary 62Gxx, secondary 62G35, 62G08, 62D20
no code implementations • 17 Jan 2020 • Jonas Peters, Stefan Bauer, Niklas Pfister
In this chapter, we provide a natural and straight-forward extension of this concept to dynamical systems, focusing on continuous time models.
Methodology Dynamical Systems
1 code implementation • 5 Nov 2019 • Niklas Pfister, Evan G. Williams, Jonas Peters, Ruedi Aebersold, Peter Bühlmann
In particular, it is useful to distinguish between stable and unstable predictors, i. e., predictors which have a fixed or a changing functional dependence on the response, respectively.
Methodology Applications
no code implementations • 28 Oct 2018 • Niklas Pfister, Stefan Bauer, Jonas Peters
Results on both simulated and real-world examples suggest that learning the structure of kinetic systems benefits from a causal perspective.
3 code implementations • 4 Jun 2018 • Niklas Pfister, Sebastian Weichwald, Peter Bühlmann, Bernhard Schölkopf
We introduce coroICA, confounding-robust independent component analysis, a novel ICA algorithm which decomposes linearly mixed multivariate observations into independent components that are corrupted (and rendered dependent) by hidden group-wise stationary confounding.
no code implementations • 1 Mar 2016 • Niklas Pfister, Peter Bühlmann, Bernhard Schölkopf, Jonas Peters
Based on an empirical estimate of dHSIC, we define three different non-parametric hypothesis tests: a permutation test, a bootstrap test and a test based on a Gamma approximation.