首页|Findings from Sage Bionetworks Update Understanding of Machine Learning (Causali ty-aware Predictions In Static Anticausal Machine Learning Tasks)
Findings from Sage Bionetworks Update Understanding of Machine Learning (Causali ty-aware Predictions In Static Anticausal Machine Learning Tasks)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Data detailed on Machine Learning have been presented. According to news reporting out of Seattle, Washington, by News Rx editors, the research stated, "We propose a counterfactual approach to train ‘causality-aware' predictive models that are able to leverage causal information in static anticausal machine learning tasks (i.e., prediction tasks where the o utcome influences the inputs). In applications plagued by confounding, the appro ach can be used to generate predictions that are free from the influence of obse rved confounders." Our news journalists obtained a quote from the research from Sage Bionetworks, " In applications involving observed mediators, the approach can be used to genera te predictions that only capture the direct or the indirect causal influences. M echanistically, we train supervised learners on (counterfactually) simulated inp uts that retain only the associations generated by the causal relations of inter est. We focus on linear models, where analytical results connecting covariances, causal effects, and prediction mean square errors are readily available. Quite importantly, we show that our approach does not require knowledge of the full ca usal graph. It suffices to know which variables represent potential confounders and/or mediators. We investigate the stability of the method with respect to dat aset shifts generated by selection biases and also relax the linearity assumptio n by extending the approach to additive models better able to account for nonlin earities in the data."
SeattleWashingtonUnited StatesNort h and Central AmericaCyborgsEmerging TechnologiesMachine LearningSage Bi onetworks
NETL
NSTL
