Maroney Lecture #1

1) In the first episode of Laurence Maroney’s, Machine Learning, Maroney expands on the difference between traditional programming and machine learning. He explains that with traditional programming, The user will input data and rules (which are made up of code) and the output will be in the form of answers (code). With machine learning, the user will input answers and data, and the program will be able to figure out the rules. This is different from traditional programming, in that with machine learning, the computer/program will be able to define the rules as the output, while traditional programming requires the user to enter the rules as an input.

2) The first time I ran the predict line with value ‘7’, I received an output of 21.997166. The second time I ran the predict line with the same value, I received an output value of 21.997938 (numbers are slightly different). The neural network begins by creating a guess which then goes to the loss to see how much error was in that guess. This result then goes to the optimizer where the optimizer will produce another guess. Following this, the optimizer value will go back to the loss and determine the amount of error in that guess. This loop continues to go on: loss to optimizer and so on. Each time the loop runs, a different guess is created. Because of this, it is possible for the values to be different depending on how much error the loss deems from the guess.

3) After fitting the new houses to the model, the model predicted 1.744 for 2 bedrooms, 2.383 for 3 bedrooms, 2.994 for 4 bedrooms, and 3.577 for 5 bedrooms. Based on the models predictions, it seems that the best deal was the Hudgins house (3 bedroom house costing $97,000). It seems that the worst deal was the Church house (with 4 bedrooms costing $399,000).