I'll give you a *very loose* **analogy** (emphasis is important here) that may help you understand the intuition. There's this technical drawing tool, called a [French curve][1], here's an example: [![enter image description here][2]][2] We were trained to use it in high school in a technical drawing class. These days, the same class is taught with CAD software, so you may not have encountered it. See, how to use them in [this video][3]. Here's a straight ruler: [![enter image description here][4]][4] <sub>(source: [officeworks.com.au](https://images.officeworks.com.au/api/2/img/https://s3-ap-southeast-2.amazonaws.com/wc-prod-pim/JPEG_300x300/SM388590_studymate_wooden_ruler_30cm.jpg/resize?size=300&auth=MjA5OTcwODkwMg__))</sub> Can you draw a curved line with a straight ruler? Of course, you can! However, it's more work. Take a look at [this video][5] to appreciate the difference. It's more efficient to use a French curve to draw curved lines than with a straight ruler. You'd have to make a lot of small lines to draft any smooth curve with the latter. [![enter image description here][6]][6] [![enter image description here][7]][7] It's not exactly the same with machine learning, but this analogy provides you with an intuition why nonlinear activation may work better in many cases: your problems are nonlinear, and having nonlinear pieces can be more efficient when combining them into a solution to nonlinear problems. [1]: https://en.wikipedia.org/wiki/French_curve [2]: https://upload.wikimedia.org/wikipedia/commons/thumb/0/0c/Krzywiki.jpg/130px-Krzywiki.jpg [3]: https://www.youtube.com/watch?v=wgATTmC0JSQ [4]: https://i.sstatic.net/qz1E3.jpg [5]: https://www.youtube.com/watch?t=2m58s&v=uulfOWzIe9o [6]: https://www.sewalongs.com/media/2011/08/using-a-French-curve-500x3331.jpg [7]: https://i.sstatic.net/pDJrB.jpg