The AI community has been a very rapidly growing community producing a vast amount of research in a very short span of time. These researches generate a lot of new methods and terminologies. With this scale of developments, it is very difficult to keep a track of terminologies, and under such conditions even more difficult to standardize the definitions. In the wake of such scenarios, we try to perform a detailed study of some terminologies in representation learning and form standard definitions of them. With this work, we establish a clear distinction between the concepts of Self-supervised learning and Weakly supervised learning.