The correct answer is (c) both LMS error & gradient descent learning law
To explain: weight update rule minimizes the mean squared error(delta square), averaged over all inputs & this laws is derived using negative gradient of error surface weight space, hence option a & b.