christinaheinze
I am a Research Scientist at Apple Health AI. My research interests include causality, distributional robustness and deep representation learning.
Pinned Repositories
backShift
Code for 'backShift', an algorithm to estimate the connectivity matrix of a directed (possibly cyclic) graph with hidden variables.
causality-course-ethz
Code used in the causality course (401-4632-15) at ETH Zurich.
christinaheinze.github.io
CompareCausalNetworks
A unified interface for the estimation of causal networks
core
TensorFlow implementation of 'Core', proposed in "Conditional Variance Penalties and Domain Shift Robustness".
latent-linear-adjustment-autoencoders
Latent Linear Adjustment autoencoders: A novel method for estimating and simulating dynamic precipitation at high resolution
learning-blackjack
Seminar in Statistics: Learning Blackjack (seminar taught at ETH Zurich in spring 2016)
loco-lib
A library for distributed L2-penalised estimation problems, implemented in Scala&Spark.
mvs-autoencoders
Code for (variational) autoencoders, provided for the lecture Multivariate Statistics at ETH Zurich in spring 2017.
nonlinearICP-and-CondIndTests
Code for a variety of nonlinear conditional independence tests and 'nonlinear Invariant Causal Prediction' to estimate the causal parents of a given target variable from data collected in different experimental or environmental conditions, extending 'Invariant Causal Prediction' from Peters, Buehlmann and Meinshausen (2016) to nonlinear settings.
christinaheinze's Repositories
christinaheinze/CompareCausalNetworks
A unified interface for the estimation of causal networks
christinaheinze/causality-course-ethz
Code used in the causality course (401-4632-15) at ETH Zurich.
christinaheinze/nonlinearICP-and-CondIndTests
Code for a variety of nonlinear conditional independence tests and 'nonlinear Invariant Causal Prediction' to estimate the causal parents of a given target variable from data collected in different experimental or environmental conditions, extending 'Invariant Causal Prediction' from Peters, Buehlmann and Meinshausen (2016) to nonlinear settings.
christinaheinze/core
TensorFlow implementation of 'Core', proposed in "Conditional Variance Penalties and Domain Shift Robustness".
christinaheinze/backShift
Code for 'backShift', an algorithm to estimate the connectivity matrix of a directed (possibly cyclic) graph with hidden variables.
christinaheinze/latent-linear-adjustment-autoencoders
Latent Linear Adjustment autoencoders: A novel method for estimating and simulating dynamic precipitation at high resolution
christinaheinze/loco-lib
A library for distributed L2-penalised estimation problems, implemented in Scala&Spark.
christinaheinze/mvs-autoencoders
Code for (variational) autoencoders, provided for the lecture Multivariate Statistics at ETH Zurich in spring 2017.
christinaheinze/learning-blackjack
Seminar in Statistics: Learning Blackjack (seminar taught at ETH Zurich in spring 2016)
christinaheinze/christinaheinze.github.io
christinaheinze/vagan-code
Tensorflow implementation of 'Visual Feature Attribution using Wasserstein GANs'