首页 > TAG信息列表 > ERM
斯坦福机器学习第九讲--经验风险最小化
本讲内容 1. Bias/Variance trade-off (偏差-方差权衡) 2. Empirical risk minimization(ERM) (经验风险最小化) 3. Union Bound/ Hoeffding inequality (联合界/霍夫丁不等式) 4. Uniform convergence (一致收敛) 1. 偏差方差权衡 对于上图左的情况,我们称之为欠拟合(under-fittingRecommendations as Treatments: Debiasing Learning and Evaluation
目录概符号说明MNAR 带来的 biasIPS EstimatorIPS Estimator 的变化性例子利用 IPS estimator 进行训练泛化界例子估计 Propensity Score泛化界疑问代码 Schnabel T., Swaminathan A., Singh A., Chandak N., Joachims T. Recommendations as treatments: debiasing learning and