Suppose we have a prediction problem where the target t corresponds to an angle, measured in...

Transcribed Image Text:

Suppose we have a prediction problem where the target t corresponds to an angle, measured in radians. A reasonable loss function we might use is: L(y, t) 1-cos(y-t) Suppose we make predictions using a linear model y = =wx+b. Part (a) - 2 pts Compute. Show your work. Part (b)- 3 pts Compute. Show your work. Part (c) 2 pts Compute DE where = L(y), t)). Show your work. 1 Part (d) - 6 pts Derive a sequence of vectorized mathematical expressions for the gradients of the cost with respect to w. As usual, the inputs are organized into a matrix X with one row per training example. The expressions should be something you can translate into a Python program without requiring a for-loop. Your answer should look like: θε Əw Suppose we have a prediction problem where the target t corresponds to an angle, measured in radians. A reasonable loss function we might use is: L(y, t) 1-cos(y-t) Suppose we make predictions using a linear model y = =wx+b. Part (a) - 2 pts Compute. Show your work. Part (b)- 3 pts Compute. Show your work. Part (c) 2 pts Compute DE where = L(y), t)). Show your work. 1 Part (d) - 6 pts Derive a sequence of vectorized mathematical expressions for the gradients of the cost with respect to w. As usual, the inputs are organized into a matrix X with one row per training example. The expressions should be something you can translate into a Python program without requiring a for-loop. Your answer should look like: θε Əw

Answer rating: 100% (QA)

Page 3 of 3 de dw ... View the full answer