Applications of Rienforcement Learning in real world

Assignment - 1

• Without using numpy (inbuilt function) write a python function to do matrix multiplication, matrix addition and calculating matrix inverse (use for loops). And then you can check your result by comparing to the values calculated by using the numpy inbuilt functions.

Assignment - 2

• You need to write Code for multi-variable linear regression Functions of your code should be easily readable and reusable.

• Given a data set attached with this message (CarPrice_Assignment.csv) . The data set is about predicting the car price based on features given(such as height, width, power, etc.) Only use following features to train your model

    carlength
    carwidth
    carheight
    enginesize
    stroke
    horsepower
    price
  

• Split the data set into training and testing data set in 80:20 ratio. Train your model with the training data set and report the accuracy of your model based on testing data set.

Assignment - 3

• The assignment is based on classifications

• On April 10, 1912, the RMS Titanic sank after colliding with an iceberg on its maiden voyage. The accident killed 1502 of the 2224 passengers and crew on board.

• The dataset has the response: whether a passenger survived the tragedy (Survived). Additional information includes the sex of the passenger (1 if male, 0 otherwise)(Sexmale), age (Age), the number of siblings/spouse aboard (SibSp), the number of parents/children aboard (Parch) and the passengers fare (Fare) (in British pounds). The training data-set contains around 880 observations. Train your classification model using this data (train.csv) and calculate percentage accuracy of your model on test data set (test.csv).

• Here's a bit information about data sets -

• train.csv - This data set is labeled with label name survived. You need train your model using label "survivied"

• test.csv - You need make predictions using this data set.

• gender_submission.csv - Compare the labels with the above predictions to calculate the accuracy.

Assignment - 4

• The main assignment is seminar_vi.ipynb. It has no requirements besides the most basic data science libraries (e.g. numpy) so you should be able to run it locally.

Assignment - 5

• start with homework.ipynb
• For seminar, implement q-learning agent and test it on Taxi and CartPole with binarizer. And then, implement EV-SARSA agent, experience replay