A computer program is said to learn from experience E w. r. t. some task T some performance measure P if its performance as measured by P oves with experience E. Suppose we feed a learning algorithm lot historical weather data and have it learn to predict weather. What would and of be a reasonable choice for Pi b) The amount of rainfall is measured in inches. Suppose you use the learning algorithm to predict how much rain will fall next week. How would you treat this classification or a regression problem? c) The exchange rate between Us$ and Japanese Yen fluctuates every day. You would like to predict whether US$ will go up or down w.r.t. yen tomorrow. How would you treat this classification or a regression problem? d) Suppose you are working on stock market prediction in terms of the number shares of NETFLIX traded daily. Millions of shares of NETFLIX are traded daily. How would you treat this classification or a regression problem? e) Suppose you ran logistic regression twice once with gamma-1 and gamma-0. one time you got w'-[25.93 56.15] and the other time w'-13.56 1.57] but you forgot which value of w' corresponds to which value of gamma, which one you think corresponds to gamma-1. 2018/5/17 17:30