EconBase
← All papers

Adversarial Generalized Method of Moments

Greg Lewis, Vasilis Syrgkanis

arXiv 19 Mar 2018 · Econometrics · 17 citations (OpenAlex)

arXiv:1803.07164 · PDF · DOI · OpenAlex · Extracted main text

Abstract

We provide an approach for learning deep neural net representations of models described via conditional moment restrictions. Conditional moment restrictions are widely used, as they are the language by which social scientists describe the assumptions they make to enable causal inference. We formulate the problem of estimating the underling model as a zero-sum game between a modeler and an adversary and apply adversarial training. Our approach is similar in nature to Generative Adversarial Networks (GAN), though here the modeler is learning a representation of a function that satisfies a continuum of moment conditions and the adversary is identifying violating moments. We outline ways of constructing effective adversaries in practice, including kernels centered by k-means clustering, and random forests. We examine the practical performance of our approach in the setting of non-parametric instrumental variable regression.

Citation extraction

20
references
26
in-text mentions
20
distinct cited
2
self-citations
20,556
main-text words

appendix boundary found by none_found · 100% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Arjovsky, Martin, Chintala, Soumith, and Bottou, Léon (2017) Wasserstein gan0.73732100%
2Rakhlin, Alexander and Sridharan, Karthik (2013) Optimization, learning, and games with predictable sequences0.64422100%
3Shalev-Shwartz, Shai Online learning and online convex optimization0.64422100%
4Goodfellow, Ian, Pouget-Abadie, Jean, Mirza, Mehdi, Xu, Bing, Warde-… (2014) Generative adversarial nets0.51121100%
5Kocaoglu, M., Snyder, C., Dimakis, A. G., and Vishwanath, S (2017) CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training0.51121100%
6Anthony, Martin and Bartlett, Peter L (2009) Neural Network Learning: Theoretical Foundations0.40511100%
7Athey, Susan, Tibshirani, Julie, and Wager, Stefan (2016) Generalized random forests0.40511100%
8Bartlett, Peter L. and Mendelson, Shahar Rademacher and gaussian complexities: Risk bounds and structural results0.40511100%
9Chamberlain, Gary (1987) Asymptotic efficiency in estimation with conditional moment restrictions0.40511100%
10Chen, Xiaohong and Liao, Zhipeng (2015) Sieve semiparametric two-step gmm under weak dependence0.40511100%

Showing the top 10 of 20 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1Dual Instrumental Variable Regression0.64422
2Minimax Instrumental Variable Regression and $L_2$ Convergence Guarantees without Identification or Closedness0.51121
3Inference on Strongly Identified Functionals of Weakly Identified Functions0.40511
4Estimating Parameters of Structural Models Using Neural Networks0.40511
5Pre-Training Estimators for Structural Models: Application to Consumer Search0.40511
6Enhancing the Merger Simulation Toolkit with ML/AI0.40511
7Epsilon-Minimax Solutions of Statistical Decision Problems0.40511
8Distributionally Robust Instrumental Variables Estimation0.00011