Tivadar Danka Profile picture
Oct 5, 2021 10 tweets 4 min read Read on X
There is one big reason we love the logarithm function in machine learning.

Logarithms help us reduce complexity by turning multiplication into addition. You might not know it, but they are behind a lot of things in machine learning.

Here is the entire story.

🧵 👇🏽
First, let's start with the definition of the logarithm.

The base 𝑎 logarithm of 𝑏 is simply the solution of the equation 𝑎ˣ = 𝑏.

Despite its simplicity, it has many useful properties that we take advantage of all the time.
You can think of the logarithm as the inverse of exponentiation.

Because of this, it turns multiplication into addition. Exponentiation does the opposite: it turns addition into multiplication.

(The base is often assumed to be a fixed constant. Thus, it can be omitted.)
Why is this useful? For calculating gradients and derivatives!

Training a neural network requires finding its gradient. However, lots of commonly used functions are written in terms of products.

As you can see, this complicates things.
By taking the logarithm, we can compute the derivative as it turns products into sums.

This method is called logarithmic differentiation.

Since the logarithm is increasing, maximizing a function is the same as maximizing its logarithm. (Same with minimization.)
One example where this is useful is the maximum likelihood estimation.

Given a set of observations and a predictive model, we can write this in the following form.
Believe it or not, this is behind the mean squared error.

Every time you use this, logarithms are working in the background.
If you enjoyed this thread and want to see behind the curtain of machine learning, I am writing a book for you, where we go from high school math to neural networks, one step at a time.

The early access for Mathematics of Machine Learning is out now!

tivadar.gumroad.com/l/mathematics-…
More applications of logarithms: transforming data for visualization. This is extremely useful in life sciences, where the scale of features is exponential.

And one more: numerical stability. When a computation involves very large numbers, taking the log can help avoid overflow.

• • •

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 3
Behold one of the mightiest tools in mathematics: the camel principle.

I am dead serious. Deep down, this tiny rule is the cog in many methods. Ones that you use every day.

Here is what it is, how it works, and why it is essential. Image
First, the story.

The old Arab passes away, leaving half of his fortune to his eldest son, third to his middle son, and ninth to his smallest.

Upon opening the stable, they realize that the old man had 17 camels. Image
This is a problem, as they cannot split 17 camels into 1/2, 1/3, and 1/9 without cutting some in half.

So, they turn to the wise neighbor for advice. Image
Read 18 tweets
Jul 3
The single biggest argument about statistics: is probability frequentist or Bayesian?

It's neither, and I'll explain why.

Buckle up. Deep-dive explanation incoming. Image
First, let's look at what is probability.

Probability quantitatively measures the likelihood of events, like rolling six with a dice. It's a number between zero and one.

This is independent of interpretation; it’s a rule set in stone. Image
In the language of probability theory, the events are formalized by sets within an event space.

The event space is also a set, usually denoted by Ω.) Image
Read 33 tweets
Jul 2
Matrix multiplication is not easy to understand.

Even looking at the definition used to make me sweat, let alone trying to comprehend the pattern. Yet, there is a stunningly simple explanation behind it.

Let's pull back the curtain! Image
First, the raw definition.

This is how the product of A and B is given. Not the easiest (or most pleasant) to look at.

We are going to unwrap this. Image
Here is a quick visualization before the technical details.

The element in the i-th row and j-th column of AB is the dot product of A's i-th row and B's j-th column. Image
Read 16 tweets
Jul 1
The single most undervalued fact of linear algebra: matrices are graphs, and graphs are matrices.

Encoding matrices as graphs is a cheat code, making complex behavior simple to study.

Let me show you how! Image
If you looked at the example above, you probably figured out the rule.

Each row is a node, and each element represents a directed and weighted edge. Edges of zero elements are omitted.

The element in the 𝑖-th row and 𝑗-th column corresponds to an edge going from 𝑖 to 𝑗.
To unwrap the definition a bit, let's check the first row, which corresponds to the edges outgoing from the first node. Image
Read 18 tweets
Jun 30
In calculus, going from a single variable to millions of variables is hard.

Understanding the three main types of functions helps make sense of multivariable calculus.

Surprisingly, they share a deep connection. Let's see why! Image
In general, a function assigns elements of one set to another.

This is too abstract for most engineering applications. Let's zoom in a little! Image
As our measurements are often real numbers, we prefer functions that operate on real vectors or scalars.

There are three categories:

1. vector-scalar,
2. vector-vector,
3. and scalar-vector. Image
Read 16 tweets
Jun 30
Neural networks are stunningly powerful.

This is old news: deep learning is state-of-the-art in many fields, like computer vision and natural language processing. (But not everywhere.)

Why are neural networks so effective? I'll explain. Image
First, let's formulate the classical supervised learning task!

Suppose that we have a dataset D, where xₖ is a data point and yₖ is the ground truth. Image
The task is simply to find a function g(x) for which

• g(xₖ) is approximately yₖ,
• and g(x) is computationally feasible.

To achieve this, we fix a parametrized family of functions. For instance, linear regression uses this function family: Image
Read 19 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!

:(