ML Algorithms

The assignment is set in three parts- use a report structure and show all how ML algorithms are used with appropriate pseudocode examples.   

Question 1. Using appropriate examples demonstrate how bias/variance tradeoffs are critical when selecting and applying ML techniques. In the report, demonstrate how overfitting problems can be mitigated using regularisation. Also provide a critique of the   role of cost functions and gradient descent methods.

Question 2. Demonstrate how ensemble techniques are employed in ML applications with bagging and boosting techniques helping to deliver better outcomes.

Question 3.  Deep learning is an emerging and promising technology for financial data modelling. Introduce Neural Networks (NNs) and explain how they may be used to solve problems in finance. You may wish to provide a working example of a Long-Short Term Memory (LSTM) model to predict stock prices. Use simple Pseudo-code to illustrate the way you will implement this technique and assess its suitability.