Yorkies, Go!

Education materials shared by the owner.

image

Blog

Cosine Similarity

![Two arrows from the origin: blue A(6,2) low and long, orange B(2,6) steep, green arc theta between them labeled cos theta = 0.60. Card reads cos theta = A.B / (|A||B|) = (6x2 + 2x6) / (sqrt(40)…

Read more

Euclidean Distance

![From O1 (1,1) to O2 (4,5): dashed legs go 3 right and 4 up, right-angle marker at the corner, and the blue straight line connecting them is labeled d = 5. Card reads d(O1,O2) = sqrt((x2-x1)^2 +…

Read more

Manhattan Distance

![Grid from O1 (1,1) to O2 (4,5): blue horizontal leg of 3 blocks, green vertical leg of 4 blocks, so the walk along the grid is 7 blocks while the straight line measures 5; card reads d = |x2 - x1|…

Read more

Leaky ReLu

# The tiny leak that saved ReLU's day ![Leaky ReLU against ReLU: negatives pass at 0.2 slope, positives untouched; card reads LeakyReLU(x) = max(x, 0.2x), x = -5 gives -1 and x = 5 gives…

Read more

LSTM vs GRU vs KDA: Is memory a dictionary?

I keep a plain explanation of [LSTM vs GRU](/post/rnn-what-is-the-difference-between-lstm-and-gru) on this site already. This is its playable sibling, plus a third contestant most explainers skip:…

Read more

Emerging AI Weekly: November 6, 2026

What's new in AI this week: shopping agents went multimodal and in-store, frontier models turned to cybersecurity at scale, image generation got faster, and regulators opened a personalized-pricing…

Read more

Loss Functions: Which to use?

*This is a series where I answer the questions my tutoring students really ask -- the ones textbooks tend to skip. I teach machine learning, deep learning, statistics, and math to learners of all…

Read more

SVM: How does Support Vector Machine work?

*This is a series where I answer the questions my tutoring students really ask -- the ones textbooks tend to skip. I teach machine learning, deep learning, statistics, and math of learners of all…

Read more

Activation Functions: Which to use?

*This is a series where I answer the questions my tutoring students really ask -- the ones textbooks tend to skip. I teach machine learning, deep learning, statistics, and math to graduate students…

Read more

Transformer Attention: How does ROPE work?

*This is a series where I answer the questions my tutoring students really ask -- the ones textbooks tend to skip. I teach machine learning, deep learning, statistics, and math to graduate students…

Read more

Emerging AI Weekly - October 30, 2026

# Emerging AI Weekly — October 30, 2026 *What's new in AI this week: world models, open weights, local AI hardware, enterprise agents, and a $17B grocery deal.* **Emerging Signals:** 1. **Google…

Read more