Tina HuynhTable of Contents
Calculus is one of the core mathematical concepts in machine learning. It plays an integral role in understanding the workings of machine learning algorithms while providing the necessary tools to optimize complex functions.
There are so many uses of calculus in machine learning - gradient computations, numerical optimization, bayesian methods, variational inference, generative adversarial networks, and more!
Learning Calculus
Learning calculus on your own doesn't have to be a strenuous road. There are many courses and resources online starting with calc 1 all the way to linear algebra, such as IXL.
Calculus as a subject in itself has an unfair reputation as being "too hard" or "not useful enough" to bother learning. In reality, calculus is a tool that allows you to solve various problems that would be much harder to solve, sometimes impossible to, without any knowledge of calculus.
In order to learn calculus, similar to many other subjects, you will want to avoid burnout as much as you possibly can. To do this, setting a schedule to remain consistent and focused in your learning all while including a minimum or maximum amount of hours will do the trick. For example:
- Monday: 6 am to 5 pm with 20 minute breaks every 2 hours - Watch lectures at 2x speed
- Tuesday: 6 am to 3 pm - Do practice problems
- Wednesday: 6 am to 5 pm with 10 minute breaks every hour - Review any questions written down for practice problems
- Thursday 6 am to 5 pm with 30 minute breaks every 2 hours - Watch more lectures at 2x speed
- Friday 6 am to 12 pm - Redo practice problems
By allowing yourself the nights, you are able to relax, go to the gym, and spend time with friends and family. If you are doing this on top of school and/or work, maybe you can shift the studying periods to night time so it is outside of work hours or on the weekends. Either way, always remember to leave room in your schedule for downtime. Your mind cannot work properly if it is always on "go" mode.
