This method is called "ridge regression". In this module, you will explore a very simple, but extremely effective technique for automatically coping with this issue. You have examined how the performance of a model varies with increasing model complexity, and can describe the potential pitfall of complex models becoming overfit to the training data. The concepts described in this module are key to all machine learning problems, well-beyond the regression setting addressed in this course. Finally, you will devise a method to first select amongst models and then assess the performance of the selected model. This leads directly to an important conversation about the bias-variance tradeoff, which is fundamental to machine learning. For these measures of error, you will analyze how they vary with model complexity and how they might be utilized to form a valid assessment of predictive performance. You will first explore the concept of measuring the "loss" of your predictions, and use this to define training, test, and generalization error. You will examine both theoretical and practical aspects of such analyses. This module is all about these important topics of model selection and assessment.
You are also ready to select amongst possible models to choose the best performing. Having learned about linear regression models and algorithms for estimating the parameters of such models, you are now ready to assess how well your considered method should perform in predicting new data. Video 25 videos A case study in predicting house prices 1m Regression fundamentals: data & model 8m Regression fundamentals: the task 2m Regression ML block diagram 4m The simple linear regression model 2m The cost of using a given line 6m Using the fitted line 6m Interpreting the fitted line 6m Defining our least squares optimization objective 3m Finding maxima or minima analytically 7m Maximizing a 1d function: a worked example 2m Finding the max via hill climbing 6m Finding the min via hill descent 3m Choosing stepsize and convergence criteria 6m Gradients: derivatives in multiple dimensions 5m Gradient descent: multidimensional hill descent 6m Computing the gradient of RSS 7m Approach 1: closed-form solution 5m Approach 2: gradient descent 7m Comparing the approaches 1m Influence of high leverage points: exploring the data 4m Influence of high leverage points: removing Center City 7m Influence of high leverage points: removing high-end towns 3m Asymmetric cost functions 3m A brief recap 1m Build a regression model to predict prices using a housing dataset. Deploy methods to select between models. Describe the notion of sparsity and how LASSO leads to sparse solutions. Estimate model parameters using optimization algorithms. Compare and contrast bias and variance when modeling data. Describe the input and output of a regression model. Learning Outcomes: By the end of this course, you will be able to: To fit these models, you will implement optimization algorithms that scale to large datasets. You will also analyze the impact of aspects of your data - such as outliers - on your selected models and predictions. You will be able to handle very large sets of features and select between models of various complexity. In this course, you will explore regularized linear regression models for the task of prediction and feature selection. Other applications range from predicting health outcomes in medicine, stock prices in finance, and power usage in high-performance computing, to analyzing which regulators are important for gene expression. This is just one of the many places where regression can be applied. In our first case study, predicting house prices, you will create models that predict a continuous value (price) from input features (square footage, number of bedrooms and bathrooms.).
0 kommentar(er)
