Corpus ID: 219573236

Deep Structural Causal Models for Tractable Counterfactual Inference

@article{Pawlowski2020DeepSC,
  title={Deep Structural Causal Models for Tractable Counterfactual Inference},
  author={Nick Pawlowski and Daniel Coelho de Castro and Ben Glocker},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.06485}
}
We formulate a general framework for building structural causal models (SCMs) with deep learning components. The proposed approach employs normalising flows and variational inference to enable tractable inference of exogenous noise variables - a crucial step for counterfactual inference that is missing from existing deep causal learning methods. Our framework is validated on a synthetic dataset built on MNIST as well as on a real-world medical dataset of brain MRI scans. Our experimental… Expand
Efficient Causal Inference from Combined Observational and Interventional Data through Causal Reductions
TLDR
Adding observational data may help to more accurately estimate causal effects even in the presence of unobserved confounders, and is found that it can often substantially reduce the number of interventional samples when adding observational training samples without sacrificing accuracy. Expand
Causal Autoregressive Flows
TLDR
This work highlights an intrinsic correspondence between a simple family of flows and identifiable causal models, and derives a bivariate measure of causal direction based on likelihood ratios, leveraging the fact that flow models estimate normalized log-densities of data. Expand
Counterfactual Generation and Fairness Evaluation Using Adversarially Learned Inference
TLDR
The proposed approach learns causal relationships between the specified attributes of an image and generates counterfactuals in accordance with these relationships that can change specified attributes and their causal descendants while keeping other attributes constant on Morpho-MNIST and CelebA datasets. Expand
Quantifying intrinsic causal contributions via structure preserving interventions
We introduce a concept to quantify the ’intrinsic’ causal contribution of each variable in a causal directed acyclic graph to the uncertainty or information of some target variable. By recursivelyExpand
Learning to synthesise the ageing brain without longitudinal data
TLDR
A deep learning-based method that learns to simulate subject-specific brain ageing trajectories without relying on longitudinal data and shows an ability to separate age, disease influence and anatomy using only 2D cross-sectional data that should be useful in large studies into neurodegenerative disease. Expand
A Structural Causal Model for MR Images of Multiple Sclerosis
TLDR
An SCM is developed that models the interaction between demographic information, disease covariates, and magnetic resonance images of the brain for people with multiple sclerosis and generates counterfactual images that show what an MR image of thebrain would look like if demographic or disease covariate are changed. Expand
A Critical Look At The Identifiability of Causal Effects with Deep Latent Variable Models
TLDR
This work investigates the gap between theory and empirical results with theoretical considerations and extensive experiments under multiple synthetic and real-world data sets, using the causal effect variational autoencoder (CEVAE) as a case study. Expand
A Tutorial on Learning Disentangled Representations in the Imaging Domain
TLDR
This tutorial paper offers an overview of the disentangled representation learning, its building blocks and criteria, and discusses applications in computer vision and medical imaging, and presents the identified opportunities for the integration of recent machine learning advances into disentanglement. Expand
Challenges for machine learning in clinical translation of big data imaging studies
TLDR
Issues relating to data availability, interpretability, evaluation and logistical challenges, and the challenges still to be overcome are focused on to enable the full success of big data deep learning approaches to be experienced outside of the research field. Expand
Counterfactual Explanations as Interventions in Latent Space
TLDR
This paper presents Counterfactual Explanations as Interventions in Latent Space (CEILS), a methodology to generate counterfactual explanations capturing by design the underlying causal relations from the data, and at the same time to provide feasible recommendations to reach the proposed profile. Expand
...
1
2
...

References

SHOWING 1-10 OF 72 REFERENCES
Deep IV: A Flexible Approach for Counterfactual Prediction
TLDR
This paper provides a recipe for augmenting deep learning methods to accurately characterize causal relationships in the presence of instrument variables (IVs)—sources of treatment randomization that are conditionally independent from the outcomes. Expand
CausalVAE: Disentangled Representation Learning via Neural Structural Causal Models
TLDR
This work proposes a new VAE based framework named CausalVAE, which includes a Causal Layer to transform independent exogenous factors into causal endogenous ones that correspond to causally related concepts in data. Expand
Causal Effect Inference with Deep Latent-Variable Models
TLDR
This work builds on recent advances in latent variable modeling to simultaneously estimate the unknown latent space summarizing the confounders and the causal effect and shows its method is significantly more robust than existing methods, and matches the state-of-the-art on previous benchmarks focused on individual treatment effects. Expand
CausalVAE: Structured Causal Disentanglement in Variational Autoencoder
TLDR
A new VAE based framework named CausalVAE is proposed, which includes causal layers to transform independent factors into causal factors that correspond to causally related concepts in data, and allows causal intervention, by which it can intervene any causal concepts to generate artificial data. Expand
Implicit Causal Models for Genome-wide Association Studies
TLDR
The first implicit causal models, a class of causal models that leverages neural architectures with an implicit density, and an implicit causal model that adjusts for confounders by sharing strength across examples are described. Expand
Counterfactual Off-Policy Evaluation with Gumbel-Max Structural Causal Models
TLDR
An off-policy evaluation procedure for highlighting episodes where applying a reinforcement learned policy is likely to have produced a substantially different outcome than the observed policy, and a class of structural causal models for generating counterfactual trajectories in finite partially observable Markov Decision Processes (POMDPs). Expand
Preventing Failures Due to Dataset Shift: Learning Predictive Models That Transport
TLDR
It is proved that the surgery estimator finds stable relationships in strictly more scenarios than previous approaches which only consider conditional relationships, and performs competitively against entirely data-driven approaches. Expand
A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms
TLDR
This work proposes to meta-learn causal structures based on how fast a learner adapts to new distributions arising from sparse distributional changes, e.g. due to interventions, actions of agents and other sources of non-stationarities and shows that causal structures can be parameterized via continuous variables and learned end-to-end. Expand
Learning Functional Causal Models with Generative Neural Networks
We introduce a new approach to functional causal modeling from observational data, called Causal Generative Neural Networks (CGNN). CGNN leverages the power of neural networks to learn a generativeExpand
CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training
TLDR
It is shown that adversarial training can be used to learn a generative model with true observational and interventional distributions if the generator architecture is consistent with the given causal graph. Expand
...
1
2
3
4
5
...