Search Results for author:
Found 22 papers, 12 papers with code
Determining the Relevance of Features for Deep Neural Networks
Deep neural networks are tremendously successful in many applications, but end-to-end trained networks often result in hard to understand black-box classifiers or predictors.
Metrics on Markov Equivalence Classes for Evaluating Causal Discovery Algorithms
Many state-of-the-art causal discovery methods aim to generate an output graph that encodes the graphical separation and connection statements of the causal graph that underlies the data-generating process.
Invariance & Causal Representation Learning: Prospects and Limitations
In causal models, a given mechanism is assumed to be invariant to changes of other mechanisms.
Identifying Linearly-Mixed Causal Representations from Multi-Node Interventions
The task of inferring high-level causal variables from low-level observations, commonly referred to as causal representation learning, is fundamentally underconstrained.
Non-parametric Conditional Independence Testing for Mixed Continuous-Categorical Variables: A Novel Method and Numerical Evaluation
One approach computes distances by a one-hot encoding of the categorical variables, essentially treating categorical variables as discrete-numerical, while the other expresses CMI by entropy terms where the categorical variables appear as conditions only.
Projecting infinite time series graphs to finite marginal graphs using number theory
In this work, we develop a method for projecting infinite time series graphs with repetitive edges to marginal graphical models on a finite time window.
Using machine learning to understand causal relationships between urban form and travel CO2 emissions across continents
Here, we address all three gaps via causal graph discovery and explainable machine learning to detect urban form effects on intra-city car travel, based on mobility data of six cities across three continents.
Conditional Independence Testing with Heteroskedastic Data and Applications to Causal Discovery
We frame heteroskedasticity in a structural causal model framework and present an adaptation of the partial correlation CI test that works well in the presence of heteroskedastic noise, given that expert knowledge about the heteroskedastic relationships is available.
Bootstrap aggregation and confidence measures to improve time series causal discovery
It can be combined with a range of time series causal discovery methods and provides a measure of confidence for the links of the time series graphs.
Discovering Causal Relations and Equations from Data
Discovering equations, laws and principles that are invariant, robust and causal explanations of the world has been fundamental in physical sciences throughout the centuries.
Selecting Robust Features for Machine Learning Applications using Multidata Causal Discovery
We apply our framework to the statistical intensity prediction of Western Pacific Tropical Cyclones (TC), for which it is often difficult to accurately choose drivers and their dimensionality reduction (time lags, vertical levels, and area-averaging).
EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task
We frame Earth surface forecasting as the task of predicting satellite imagery conditioned on future weather.
Ranked #5 on
Earth Surface Forecasting
on EarthNet2021 OOD Track
Towards Learning an Unbiased Classifier from Biased Data via Conditional Adversarial Debiasing
Often, the bias of a classifier is a direct consequence of a bias in the training dataset, frequently caused by the co-occurrence of relevant features and irrelevant ones.
Necessary and sufficient graphical conditions for optimal adjustment sets in causal graphical models with hidden variables
In the present work optimality is characterized as maximizing a certain adjustment information which allows to derive a necessary and sufficient graphical criterion for the existence of an optimal adjustment set and a definition and algorithm to construct it.
EarthNet2021: A novel large-scale dataset and challenge for forecasting localized climate impacts
Here, we define high-resolution Earth surface forecasting as video prediction of satellite imagery conditional on mesoscale weather forecasts.
High-recall causal discovery for autocorrelated time series with latent confounders
We show that existing causal discovery methods such as FCI and variants suffer from low recall in the autocorrelated time series case and identify low effect size of conditional independence tests as the main reason.
A Perspective on Gaussian Processes for Earth Observation
Earth observation (EO) by airborne and satellite remote sensing and in-situ observations play a fundamental role in monitoring our planet.
Discovering contemporaneous and lagged causal relations in autocorrelated nonlinear time series datasets
The paper introduces a novel conditional independence (CI) based method for linear and nonlinear, lagged and contemporaneous causal discovery from observational time series in the causally sufficient case.
Conditional independence testing based on a nearest-neighbor estimator of conditional mutual information
Combining the local permutation scheme with the kernel tests leads to better calibration, but suffers in power.
Detecting causal associations in large nonlinear time series datasets
Detecting causal associations in time series datasets is a key challenge for novel insights into complex dynamical systems such as the Earth system or the human brain.
Methodology Atmospheric and Oceanic Physics Applications
Unified functional network and nonlinear time series analysis for complex systems science: The pyunicorn package
Additionally, \texttt{pyunicorn} provides insights into the nonlinear dynamics of complex systems as recorded in uni- and multivariate time series from a non-traditional perspective by means of recurrence quantification analysis (RQA), recurrence networks, visibility graphs and construction of surrogate time series.
Data Analysis, Statistics and Probability Atmospheric and Oceanic Physics
Optimal model-free prediction from multivariate time series
Here a prediction scheme that overcomes this strong limitation is introduced utilizing a causal pre-selection step which drastically reduces the number of possible predictors to the most predictive set of causal drivers making a globally optimal search scheme tractable.
