Estimating Treatment Effect in the Wild via Differentiated Confounder Balancing

Kuang, Kun; Cui, Peng; Li, Bo; Jiang, Meng; Yang, Shiqiang

doi:10.1145/3097983.3098032

Cited by 63 publications

(38 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods Standard approaches in ITE prediction include model-agnostic methods (meta-learners [38; 44], modified outcome methods [5] and their combinations [35]) that implicitly predict the CATE, as well as tree-based approaches which are particularly suited for direct treatment effect estimation [4; 55]. A prolific series of increasingly performing algorithm targeted towards ITE prediction have been proposed, using a variety of techniques to adjust for the covariate shift, such as generative adversarial nets [56], auto-encoders [43], double machine learning [15], representation learning [54; 58] or confounder balancing algorithms [37]. Another recent trend is to study theoretical limits in ITE prediction and especially generalization bounds [2].…”

Section: Individual Treatment Effectmentioning

confidence: 99%

A Large Scale Benchmark for Individual Treatment Effect Prediction and Uplift Modeling

Diemert¹,

Betlei²,

Renaudin³

et al. 2021

Preprint

View full text Add to dashboard Cite

Individual Treatment Effect (ITE) prediction is an important area of research in machine learning which aims at explaining and estimating the causal impact of an action at the granular level. It represents a problem of growing interest in multiple sectors of application such as healthcare, online advertising or socioeconomics. To foster research on this topic we release a publicly available collection of 13.9 million samples collected from several randomized control trials, scaling up previously available datasets by a healthy 210x factor. We provide details on the data collection and perform sanity checks to validate the use of this data for causal inference tasks. First, we formalize the task of uplift modeling (UM) that can be performed with this data, along with the relevant evaluation metrics. Then, we propose synthetic response surfaces and heterogeneous treatment assignment providing a general set-up for ITE prediction. Finally, we report experiments to validate key characteristics of the dataset leveraging its size to evaluate and compare − with high statistical significance − a selection of baseline UM and ITE prediction methods.

show abstract

Section: Individual Treatment Effectmentioning

confidence: 99%

A Large Scale Benchmark for Individual Treatment Effect Prediction and Uplift Modeling

Diemert¹,

Betlei²,

Renaudin³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Sample reweighting methods aim to correct the treatment assignment using observational data in order to overcome the subject selection bias. For instance, these methods include Inverse Propensity Weighting (IPW) based on propensity score [17] and confounder balancing methods [12]. Other methods include doubly-robust approaches which combine covariate adjustment with propensity score weighting [3], [4], [5].…”

Section: Related Workmentioning

confidence: 99%

Causal Effect Estimation using Variational Information Bottleneck

Cheng

Zhong

et al. 2021

Preprint

View full text Add to dashboard Cite

Causal inference is to estimate the causal effect in a causal relationship when intervention is applied. Precisely, in a causal model with binary interventions, i.e., control and treatment, the causal effect is simply the difference between the factual and counterfactual. The difficulty is that the counterfactual may never been obtained which has to be estimated and so the causal effect could only be an estimate. The key challenge for estimating the counterfactual is to identify confounders which effect both outcomes and treatments. A typical approach is to formulate causal inference as a supervised learning problem and so counterfactual could be predicted. Including linear regression and deep learning models, recent machine learning methods have been adapted to causal inference. In this paper, we propose a method to estimate Causal Effect by using Variational Information Bottleneck (CEVIB). The promising point is that VIB is able to naturally distill confounding variables from the data, which enables estimating causal effect by using observational data. We have compared CEVIB to other methods by applying them to three data sets showing that our approach achieved the best performance. We also experimentally showed the robustness of our method.

show abstract

“…In the literature of causality [23], an ideal model to resolve selection bias is to make policy based on causal variables, which keep stable across different domains [24]. Popular methods based one observational data to estimate the causal effect of a treatment on the outcome include propensity score matching [25,26], markov blankets [27,28] and confounder balancing [29,30] and etc [31]. Lately [32] leverage causality for predictive modeling.…”

Section: Proportional Bias Settingmentioning

confidence: 99%

Towards Non-I.I.D. Image Classification: A Dataset and Baselines

He¹,

Shen²,

Cui³

2019

Preprint

Self Cite

View full text Add to dashboard Cite

I.I.D. 1 hypothesis between training and testing data is the basis of numerous image classification methods. Such property can hardly be guaranteed in practice where the Non-IIDness is common, causing instable performances of these models. In literature, however, the Non-I.I.D. 2 image classification problem is largely understudied. A key reason is lacking of a well-designed dataset to support related research. In this paper, we construct and release a Non-I.I.D. image dataset called NICO 3 , which uses contexts to create Non-IIDness consciously. Compared to other datasets, extended analyses prove NICO can support various Non-I.I.D. situations with sufficient flexibility. Meanwhile, we propose a baseline model with ConvNet structure for General Non-I.I.D. image classification, where distribution of testing data is unknown but different from training data. The experimental results demonstrate that NICO can well support the training of ConvNet model from scratch, and a batch balancing module can help ConvNets to perform better in Non-I.I.D. settings.

show abstract

Estimating Treatment Effect in the Wild via Differentiated Confounder Balancing

Cited by 63 publications

References 22 publications

A Large Scale Benchmark for Individual Treatment Effect Prediction and Uplift Modeling

A Large Scale Benchmark for Individual Treatment Effect Prediction and Uplift Modeling

Causal Effect Estimation using Variational Information Bottleneck

Towards Non-I.I.D. Image Classification: A Dataset and Baselines

Contact Info

Product

Resources

About