{"id":5124,"date":"2021-01-20T08:39:14","date_gmt":"2021-01-20T08:39:14","guid":{"rendered":"https:\/\/joapen.com\/blog\/?p=5124"},"modified":"2021-01-20T08:39:14","modified_gmt":"2021-01-20T08:39:14","slug":"tuning-a-machine-learning-model","status":"publish","type":"post","link":"https:\/\/joapen.com\/blog\/2021\/01\/20\/tuning-a-machine-learning-model\/","title":{"rendered":"Tuning a Machine Learning Model"},"content":{"rendered":"

I continue taking some basic notes of the book “Enterprise Artificial Intelligence and Machine Learning for Managers<\/a>“.<\/p>\n

Tuning a machine learning model is an iterative<\/strong> process. Data scientists typically run
\nnumerous experiments to train and evaluate models, trying out different features,
\ndifferent loss functions, different AI\/ML models, and adjusting model parameters
\nand hyper-parameters.<\/p>\n

Feature engineering<\/h1>\n

Feature engineering broadly refers to mathematical transformations of raw data in order to feed appropriate signals into AI\/ML models.<\/p>\n

In real world data are derived from a variety of source systems and typically are not reconciled or aligned in time and space. Data scientists often put significant effort into defining data transformation pipelines and building out their feature vectors.<\/p>\n

In addition, data scientists should implement requirements for feature normalization or scaling to ensure that no one feature overpowers the algorithm.<\/p>\n

Loss Functions<\/h1>\n

A loss function serves as the objective function that the AI\/ML algorithm is seeking to optimize during training efforts. During model training, the AI\/ML algorithm aims to minimize the loss function. Data scientists often consider different loss functions to
\nimprove the model \u2013 e.g., make the model less sensitive to outliers, better handle noise, or reduce over-fitting.<\/p>\n

A simple example of a loss function is mean squared error (MSE), which often is used to optimize regression models. MSE measures the average of squared difference between predictions and actual output values.<\/p>\n

\"\"<\/a>
These two linear regression models have the same MSE, but the model on
the left is under-predicting and the model on the right is over-predicting.<\/figcaption><\/figure>\n

It is important, to recognize the weaknesses of loss functions. Over-relying on loss functions as an indicator of prediction accuracy may lead to erroneous model set points.<\/p>\n

Regularization<\/h1>\n

Regularization is a method to balance over-fitting and under-fitting a model during training. Both over-fitting and under-fitting are problems that ultimately cause poor predictions on new data.<\/p>\n