Various typos

[danakianfar]

A few typos I found with the lecture notes.

1. Def 5.2.3: I think w and m are being used interchangeably. \lambda_w was the notation used in class.
2. Def 5.2.12: In the following paragraph, m should be a vector of size k. Even though m is a binary vector, I think using \mathcal{X}^k is a confusing notation since \mathcal{X} is used to refer to the inputs set.
3. Example 5.2.17: G^T should be a 5x1 matrix
4. Example 6.0.1: The output set is {1,2,3}, but the encoding function is taking values in {0,1,2}
5. Def 6.1.3: Both should be conditional densities.