2016 LearningRepresentationsforCount

• (Johansson et al., 2016) ⇒ Fredrik D. Johansson, Uri Shalit, and David Sontag. (2016). “Learning Representations for Counterfactual Inference.” In: Proceedings of the 33rd International Conference on International Conference on Machine Learning - Volume 48.

Subject Headings: Counterfactual Inference.

Cited By

• http://scholar.google.com/scholar?q=%222016%22+Learning+Representations+for+Counterfactual+Inference
• http://dl.acm.org/citation.cfm?id=3045390.3045708&preflayout=flat#citedby

Quotes

Abstract

Observational studies are rising in importance due to the widespread accumulation of data in fields such as healthcare, education, employment and ecology. We consider the task of answering counterfactual questions such as, " Would this patient have lower blood sugar had she received a different medication? ". We propose a new algorithmic framework for counterfactual inference which brings together ideas from domain adaptation and representation learning. In addition to a theoretical justification, we perform an empirical comparison with previous approaches to causal inference from observational data. Our deep learning algorithm significantly outperforms the previous state-of-the-art.

…

5. Related work

Counterfactual inference for determining causal effects in observational studies has been studied extensively in statistics, economics, epidemiology and sociology (Morgan & Winship, 2014; Robins et al., 2000; Rubin, 2011; Chernozhukov et al., 2013) as well as in machine learning (Langford et al., 2011; Bottou et al., 2013; Swaminathan & Joachims, 2015a).

Non-parametric methods do not attempt to model the relation between the context, intervention, and outcome. The methods include nearest-neighbor matching, propensity score matching, and propensity score re-weighting (Rosenbaum & Rubin, 1983; Rosenbaum, 2002; Austin, 2011). Parametric methods, on the other hand, attempt to concretely model the relation between the context, intervention, and outcome. These methods include any type of regression including linear and logistic regression (Prentice, 1976; Gelman & Hill, 2006), random forests (Wager &Athey, 2015) and regression trees (Chipman et al., 2010).

Doubly robust methods combine aspects of parametric and non-parametric methods, typically by using a propensity score weighted regression (Bang & Robins, 2005; Dud´ık et al., 2011). They are especially of use when the treatment assignment probability is known, as is the case for off-policy evaluation or learning from logged bandit data. Once the treatment assignment probability has to be estimated, as is the case in most observational studies, their efficacy might wane considerably (Kang & Schafer, 2007).

Tian et al. (2014) presented one of the few methods that achieve balance by transforming or selecting covariates, modeling interactions between treatment and covariates.

…

References

;

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2016 LearningRepresentationsforCount	David Sontag Fredrik D. Johansson Uri Shalit			Learning Representations for Counterfactual Inference						2016