Following are the key points described later in this article:
When working with machine learning algorithm, the primary objective is to minimize the error between predicted and observed value by minimizing the cost function which, for regression models, is mean squared difference between predicted and observed value (added with regularization if required). Below argument can as well be applied for neural network. Following are different reasons why large prediction errors could be found while working with linear or logistic regression models.
Following are some of the techniques that one could use to reduce the prediction error and further optimize the regression models. Note that the error results from common reasons such as high bias (under-fitting) or high variance (over-fitting)
Last updated: 28th Aug, 2024 Confusion among data scientists regarding ROC Curve and AUC often…
Last updated: 27th Aug, 2024 Classification models are used in classification problems to predict the…
Last updated: 26th August, 2024 In this blog post, we will discuss the concepts of…
Last updated: 25th August, 2024 In machine learning, model complexity and overfitting are related in…
Last updated: 24th August, 2024 Model parallelism and data parallelism are two strategies used to…
Last updated: 24th August, 2024 The performance of the machine learning models on unseen datasets…