What I learned teaching Machine Learning to University Students
So This year I have done two talks for Swinburne University Vietnam related to malware analysis. But this time we are trying to keep up with the times and use machine learning. In this blog I will speak about my experience prepping and presenting this workshop
The Prep
Teaching machine learning is not an easy feat, you need to sit down and understand machine learning and how to use it and understand the coding math involved! I can understand why many people chase the AI dream but also why many fail.
So during my prep I searched for machine learning material to push the workshop along! I found EMBER, which is a dataset of malware components in different languages! Due to the sheer size of EMBER I only extracted the .net malware components and that was still roughly 2 gb's of data so definitely need to exercise caution when downloading datasets because they can get big...quickly!
Next I went ahead and developed the notebooks which were split up into 3 ideas:
- Toy data
- Hybrid data with ember
- Full Ember dataset and presentation
This was unique because the main goal was to show the math and set up behind preparing machine learning data and how the model can make a prediction. this was super interesting because this territory was very new to me. But I got to see how we can transform math into meaningful information! This was amazing because as I researched and implemented, I learned how machine learning, if done correctly, can be a huge benefit to society!
The Workshop
So after all my notebooks and code were running It was time to teach! In the classroom, I roughly had 20 students. The class size was manageable, and people were engaged! It was fun because people were curious how malware analysts determined the data for all this and had many questions! I think the best part was watching students become curious and wanting to develop their own models, and I think it's great because yes, machine learning has developed so much, but there's still room for growth, so I was quite excited to see them enjoy it!
What I learned
Overall I learned a lot about my teaching style at a group level, normally I do one on one lessons and the occasional group lessons but this was interesting because I had to keep the students engaged and learning and that is a hard concept not as a educator but as a millennial because we didn't have things like kahoot or certain games to keep us entertained so I went around and asked questions but I can definitely see kahoot being engaging so i will be improving my teaching style to add this more! But in general, the students were happy and had follow-up questions, so I am very happy about that!
P.S
I am very thankful for the opportunity to teach at Swinburne University, and it was an amazing experience to put my educational and technical background to work! So it was a great time!
Thank you for reading!
Until next time!
