Information Bottleneck for Estimating Treatment Effects with Systematically Missing Covariates

Parbhoo, Sonali; Wieser, Mario; Wieczorek, Aleksander; Röth, Volker

doi:10.3390/e22040389

Cited by 12 publications

(7 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [29] a doubly robust based method is suggested. In [30], authors consider missing values only during test time and suggest a method based on information bottleneck technique. Finally, [31] suggests a new method based on VAEs (adopted for missing values) which learns distribution of the latent confounder and hence assumes a weaker condition than unconfoundedness with missing values which is harder to justify.…”

Section: A Related Workmentioning

confidence: 99%

Conservative Policy Construction Using Variational Autoencoders for Logged Data with Missing Values

Abroshan,

Yip,

Tekin

et al. 2021

Preprint

View full text Add to dashboard Cite

In high-stakes applications of data-driven decision making like healthcare, it is of paramount importance to learn a policy that maximizes the reward while avoiding potentially dangerous actions when there is uncertainty. There are two main challenges usually associated with this problem. Firstly, learning through online exploration is not possible due to the critical nature of such applications. Therefore, we need to resort to observational datasets with no counterfactuals. Secondly, such datasets are usually imperfect, additionally cursed with missing values in the attributes of features. In this paper, we consider the problem of constructing personalized policies using logged data when there are missing values in the attributes of features in both training and test data. The goal is to recommend an action (treatment) when X, a degraded version of X with missing values, is observed. We consider three strategies for dealing with missingness. In particular, we introduce the conservative strategy where the policy is designed to safely handle the uncertainty due to missingness. In order to implement this strategy we need to estimate posterior distribution p(X| X), we use variational autoencoder to achieve this. In particular, our method is based on partial variational autoencoders (PVAE) which are designed to capture the underlying structure of features with missing values.

show abstract

Section: A Related Workmentioning

confidence: 99%

Conservative Policy Construction Using Variational Autoencoders for Logged Data with Missing Values

Abroshan,

Yip,

Tekin

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Various approaches exploit this relation, e.g the deep variational information bottleneck (DVIB) [2,4]. Further extensions were proposed in the context of causality [9,29,30] or archetypal analysis [15,16]. The β-VAE [13] extends the standard VAE approach and allows unsupervised disentanglement.…”

Section: Deep Generative Latent Variable Models and Disentanglementmentioning

confidence: 99%

Learning Conditional Invariance through Cycle Consistency

Samarin,

Nesterov,

Wieser

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Identifying meaningful and independent factors of variation in a dataset is a challenging learning task frequently addressed by means of deep latent variable models. This task can be viewed as learning symmetry transformations preserving the value of a chosen property along latent dimensions. However, existing approaches exhibit severe drawbacks in enforcing the invariance property in the latent space. We address these shortcomings with a novel approach to cycle consistency. Our method involves two separate latent subspaces for the target property and the remaining input information, respectively. In order to enforce invariance as well as sparsity in the latent space, we incorporate semantic knowledge by using cycle consistency constraints relying on property side information. The proposed method is based on the deep information bottleneck and, in contrast to other approaches, allows using continuous target properties and provides inherent model selection capabilities. We demonstrate on synthetic and molecular data that our approach identifies more meaningful factors which lead to sparser and more interpretable models with improved invariance properties.

show abstract

“…Some empirical work shows that encoder-based models with enough proxies (variables caused by hidden confounders) can improve causal inference under hidden confounding [32,36], and theoretical work proves the identifiability of latent variables from proxies under some assumptions [2,28].…”

Section: Hidden Confoundingmentioning

confidence: 99%

Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Data

Löwe¹,

Madras²,

Zemel³

et al. 2020

Preprint

View full text Add to dashboard Cite

Standard causal discovery methods must fit a new model whenever they encounter samples from a new underlying causal graph. However, these samples often share relevant information -for instance, the dynamics describing the effects of causal relations -which is lost when following this approach. We propose Amortized Causal Discovery, a novel framework that leverages such shared dynamics to learn to infer causal relations from time-series data. This enables us to train a single, amortized model that infers causal relations across samples with different underlying causal graphs, and thus makes use of the information that is shared. We demonstrate experimentally that this approach, implemented as a variational model, leads to significant improvements in causal discovery performance, and show how it can be extended to perform well under hidden confounding. * equal contribution Preprint. Under review.

show abstract

Information Bottleneck for Estimating Treatment Effects with Systematically Missing Covariates

Cited by 12 publications

References 29 publications

Conservative Policy Construction Using Variational Autoencoders for Logged Data with Missing Values

Conservative Policy Construction Using Variational Autoencoders for Logged Data with Missing Values

Learning Conditional Invariance through Cycle Consistency

Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Data

Contact Info

Product

Resources

About