Tivadar Danka Profile picture
Apr 15, 2021 9 tweets 3 min read Read on X
In machine learning, the inner product (or dot product) of vectors is often used to measure similarity.

However, the formula is far from revealing. What does the sum of coordinate products have to do with similarity?

There is a very simple geometric explanation!

🧵 👇🏽
There are two key things to observe.

First, the inner product is linear in both variables. This property is called bilinearity.
Second, is that the inner product is zero if the vectors are orthogonal.
With these, given an 𝑦, we can decompose 𝑥 into two components: one is orthogonal, while the other is parallel to 𝑦.

So, because of the bilinearity, the inner product equals to the inner product of 𝑦 and the parallel component of 𝑥.
If we write 𝑦 as a scalar multiple of 𝑥, we can see that their inner product can be expressed in terms of the magnitude of 𝑦 and the scalar.
In addition, if we assume that 𝑥 and 𝑦 have unit magnitude, the inner product is even simpler: it describes the scaling factor between 𝑦 and the orthogonal projection of 𝑥 onto 𝑦.

Note that this factor is in [-1, 1]. (It is negative if the directions are opposite.)
There is more. Now comes the really interesting part!

Let's denote the angle between 𝑥 and 𝑦 by α. The scaling factor r equals the cosine of α!

(Recall that we assume that 𝑥 and 𝑦 have unit magnitude.)
If the vectors don't have unit magnitude, we can simply scale them.

The inner product of the scaled vectors is called cosine similarity.

This is probably how you know this quantity. Now you see why!
If you have liked this thread, consider following me and hitting a like/retweet on the first tweet of the thread!

I regularly post simple explanations of seemingly complicated concepts in machine learning, make sure you don't miss out on the next one!

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Tivadar Danka

Tivadar Danka Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @TivadarDanka

Jul 14
"Probability is the logic of science."

There is a deep truth behind this conventional wisdom: probability is the mathematical extension of logic, augmenting our reasoning toolkit with the concept of uncertainty.

In-depth exploration of probabilistic thinking incoming. Image
Our journey ahead has three stops:

1. an introduction to mathematical logic,
2. a touch of elementary set theory,
3. and finally, understanding probabilistic thinking.

First things first: mathematical logic.
In logic, we work with propositions.

A proposition is a statement that is either true or false, like
• "it's raining outside",
• or "the sidewalk is wet".

These are often abbreviated as variables, such as A = "it's raining outside".
Read 28 tweets
Jul 13
Conditional probability is the single most important concept in statistics.

Why? Because without accounting for prior information, predictive models are useless.

Here is what conditional probability is, and why it is essential. Image
Conditional probability allows us to update our models by incorporating new observations.

By definition, P(B | A) describes the probability of an event B, given that A has occurred. Image
Here is an example. Suppose that among 100 emails, 30 are spam.

Based only on this information, if we inspect a random email, our best guess is a 30% chance of it being a spam.

This is not good enough. Image
Read 10 tweets
Jul 11
Most people think math is just numbers.

But after 20 years with it, I see it more like a mirror.

Here are 10 surprising lessons math taught me about life, work, and thinking clearly: Image
1. Breaking the rules is often the best course of action.

We have set theory because Bertrand Russell broke the notion that “sets are just collections of things.”
2. You have to understand the rules to successfully break them.

Miles Davis said, “Once is a mistake, twice is jazz.”

Mistakes are easy to make. Jazz is hard.
Read 12 tweets
Jul 8
This will surprise you: sine and cosine are orthogonal to each other.

What does orthogonality even mean for functions? In this thread, we'll use the superpower of abstraction to go far beyond our intuition.

We'll also revolutionize science on the way. Image
Our journey ahead has three milestones. We'll

1. generalize the concept of a vector,
2. show what angles really are,
3. and see what functions have to do with all this.

Here we go!
Let's start with vectors. On the plane, vectors are simply arrows.

The concept of angle is intuitive as well. According to Wikipedia, an angle “is the figure formed by two rays”.

How can we define this for functions? Image
Read 18 tweets
Jul 7
In machine learning, we use the dot product every day.

However, its definition is far from revealing. For instance, what does it have to do with similarity?

There is a beautiful geometric explanation behind. Image
By definition, the dot product (or inner product) of two vectors is defined by the sum of coordinate products. Image
To peek behind the curtain, there are three key properties that we have to understand.

First, the dot product is linear in both variables.

This property is called bilinearity. Image
Read 15 tweets
Jul 5
If I had to learn Math for Machine Learning from scratch, this is the roadmap I would follow: Image
1. Linear Algebra

These are non-negotiables:

• Vectors
• Matrices
• Equations
• Factorizations
• Matrices and graphs
• Linear transformations
• Eigenvalues and eigenvectors

Now you've learned how to represent and transform data. Image
2. Calculus

Don't skip any of these:

• Series
• Functions
• Sequences
• Integration
• Optimization
• Differentiation
• Limits and continuity

Now you understand the math behind algorithms like gradient descent and get a better feeling of what optimization is. Image
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(