Top Three Insights:

When I started this class, I wished someone told me why this class was a requirement for the major. I disliked probability and statistics when I took it in high school. I was surprised by how much I enjoyed this class. These are my top three takeaways from the class, and I hope others will find them useful!

1. Professor Beame’s key takeaway: the applications of the course extend beyond computer science and math.

   • The most obvious is detecting statistical lies.

     • “Torture numbers, and they’ll confess to anything”- George Easterbrook

   • Green jelly beans give you acne.

     • The quote by George Easterbrook was followed by an interesting cartoon of lying with statistics when correlating eating jelly beans and levels of acne. There were several different experiments without significant results. However, they kept repeating the experiment until they found significant results. Eventually, they did.

2. AI/ML/NLP foundations

   • Having taken an advanced AI class (link to class), I realized the importance of having a strong foundation before exploring newer, exciting concepts. Knowing probability/statistics and the intuition behind the calculations are a backbone for these algorithms that leverage basic probability calculations.

3. Deep connection between algorithms and probability

   • Some of the problem sets included programming assignments, which were a great way of transferring the knowledge from mathematical calculations to code. We learned a variety of algorithms and implemented four:

     • Naive Bayes

     • Bloom filters

     • MinHash

     • Markov’s Chains

   • Although it was not touched on in-depth, probability and statistics are crucial for runtime analysis.