Unknown

Dataset Information

0

Learning to learn from data: Using deep adversarial learning to construct optimal statistical procedures.

ABSTRACT: Traditionally, statistical procedures have been derived via analytic calculations whose validity often relies on sample size growing to infinity. We use tools from deep learning to develop a new approach, adversarial Monte Carlo meta-learning, for constructing optimal statistical procedures. Statistical problems are framed as two-player games in which Nature adversarially selects a distribution that makes it difficult for a statistician to answer the scientific question using data drawn from this distribution. The players' strategies are parameterized via neural networks, and optimal play is learned by modifying the network weights over many repetitions of the game. Given sufficient computing time, the statistician's strategy is (nearly) optimal at the finite observed sample size, rather than in the hypothetical scenario where sample size grows to infinity. In numerical experiments and data examples, this approach performs favorably compared to standard practice in point estimation, individual-level predictions, and interval estimation.

SUBMITTER: Luedtke A 

PROVIDER: S-EPMC7051830 | biostudies-literature | 2020 Feb

REPOSITORIES: biostudies-literature

Json Xml
altmetric image

Publications

Learning to learn from data: Using deep adversarial learning to construct optimal statistical procedures.

Luedtke Alex A   Carone Marco M   Simon Noah N   Sofrygin Oleg O  

Science advances 20200226 9

Traditionally, statistical procedures have been derived via analytic calculations whose validity often relies on sample size growing to infinity. We use tools from deep learning to develop a new approach, adversarial Monte Carlo meta-learning, for constructing optimal statistical procedures. Statistical problems are framed as two-player games in which Nature adversarially selects a distribution that makes it difficult for a statistician to answer the scientific question using data drawn from thi  ...[more]

Similar Datasets

Unknown Deploying deep learning models on unseen medical imaging using adversarial domain adaptation.
| S-EPMC9565422 | biostudies-literature
Unknown Adversarial Robustness of Deep Reinforcement Learning Based Dynamic Recommender Systems.
| S-EPMC9110778 | biostudies-literature
Unknown MRI restoration using edge-guided adversarial learning.
| S-EPMC7977797 | biostudies-literature
Unknown DReLAB - Deep REinforcement Learning Adversarial Botnet: A benchmark dataset for adversarial attacks against botnet Intrusion Detection Systems.
| S-EPMC7749366 | biostudies-literature
Unknown Unsupervised single-particle deep classification via statistical manifold learning
| EMPIAR-10069 | biostudies-other
Unknown A deep learning adversarial autoencoder with dynamic batching displays high performance in denoising and ordering scRNA-seq data.
| S-EPMC10867661 | biostudies-literature
Genomics Modification detection using deep learning on Nanopore data
| PRJEB31789 | ENA
Unknown Learning statistical models of phenotypes using noisy labeled training data.
| S-EPMC5070523 | biostudies-literature
Unknown Natural Images Allow Universal Adversarial Attacks on Medical Image Classification Using Deep Neural Networks with Transfer Learning.
| S-EPMC8875959 | biostudies-literature
Unknown Optimal adaptive allocation using deep reinforcement learning in a dose-response study.
| S-EPMC9298337 | biostudies-literature