SGD Update

Perform one stochastic-gradient-descent update: $w\leftarrow w-\eta gw\backslash leftarrow\; w-\backslash eta\backslash ,g$.

Implement solve(w,grad,lr).

Returns. Return the updated parameters as a NumPy array (the input is not modified).

Examples

Example 1

Input:

solve([1.0, 2.0], [0.1, 0.2], 0.01)

Output:

[0.999, 1.998]

Example 2

Input:

solve([5.0], [-2.0], 0.5)

Output:

[6.0]