Katie Codes

It’s been almost three months since my last post! That was quite a hectic time in my life, mainly down to my decision to apply for the Data Lab MSc Scholarship in July. I was awarded the scholarship in August and left my job in front end development to start the MSc in September at Heriot-Watt University. So far the experience has been brilliant, I’ve really enjoyed getting to immerse myself in study again. The course is completely online (for this semester at least), which is quite different to my previous university experiences. I’m also the class rep, which has been a nice excuse to meet more of the students and make sure everyone is doing okay. Apart from some awful family news, this would have been the perfect start to a new chapter in my life.

My courses this semester are:

1. Artificial Intelligence and Intelligent Agents
2. Biologically Inspired Computation
3. Data Mining and Machine Learning
4. Software Engineering Foundations

My favourite course so far is Biologically Inspired Computation, though I may be biased due to my biological background! I’ve spent a lot of time revising calculus, which has been really fun. We’ve covered perceptrons and have just started on the more advanced neural networks. Here are some resources I’ve found invaluable so far:

1. Khan Academy for all things maths, with particular focus on maths required for neural networks: partial differentiation, chain rule, matrix multiplication
2. 3Blue1Brown produced an excellent, accessible YouTube series on multilayer perceptrons.
3. Russell and Norvig is a classic AI textbook, I found chapter 18 (in the 3rd edition) particularly useful for studying machine learning.

I’m looking forward to learning more as the semester progresses and getting down to some coding. I’m learning Java on the Software Engineering Foundations course, which is not the most popular language for data science/machine learning applications. It’s a nice refresher however (weirdly for bioinformatics, I used Java during my PhD). I’m dusting off my Python skills for the Biologically Inspired Computation course, Python being arguably the most popular language for data science/machine learning.