DEV Community

Cover image for Learn ML in 52 seconds 🏎️
Dima Sukharev
Dima Sukharev

Posted on • Updated on

Learn ML in 52 seconds 🏎️

first lets grow my twitter, i will probably change all of the world obsolete education systems, not sure yet, the more people follow me — the easier it is to change the systems


one more speedrun in the series, let's goooo

Image description

i just used reinforcement learning (RL) meme on you:

  1. you open the post (do a right thing)
  2. you get the meme (a reward)

what a bait it was.. omg..


so, why is it hard for meme-Kelvin to learn what RL is?
Because he wants to learn the implementation of RL instead of understanding the Concept, THATS WHY meme-KELVIN!

when you learn a new thing, tool or tech or anything — you do NOT start with it's implementations, you start with ideas, concepts and problems the tool solves!

so Kelvin, here you go:

to understand Reinforcement Learning (RL) — think about playing a video game where you earn points for making the right moves.

reinforcement learning is like that - a program learns by making decisions and getting rewards or penalties based on its actions

that was just an example of applying START-WITH-IDEAS principle to Reinforcement Learning

but i promised to tell how to learn the whole ML in 52 seconds...


F&F3 drag race

THE SPEEDRUN!

to learn ML in 52 seconds you learn the ML Concepts, not implementations, then you google (or GPT) implementations in PyTorch or whatever library name you like, anyway it is going to change in the next year and it's fine, ideas will stay the same for much longer — go for ideas!

Machine Learning Ideas

1. Supervised Learning

In supervised learning, a program is taught using examples with answers (called labeled data). This helps the program learn the connection between the examples and answers, so it can guess the answers for new examples it hasn't seen before.

Algorithms and problems to solve: predicting house prices (linear regression), deciding if a customer will buy a product (decision trees)

2. Unsupervised Learning

Here, the program looks at examples without answers (unlabeled data) and tries to find patterns or groups within them. This can help with tasks like grouping similar items together or reducing the amount of information needed to describe the data.

Algorithms and problems to solve: grouping people with similar music tastes (k-means clustering), compressing images without losing too much information (principal component analysis)

3. Reinforcement Learning

In reinforcement learning, the program learns to make decisions by trying things out and getting feedback in the form of rewards or penalties. The goal is to make better decisions over time and solve problems more effectively.

Algorithms and problems to solve: teaching a robot to walk (Q-learning), training a program to play chess (policy gradient methods)

4. Feature Engineering

This is the process of choosing important information (features) from raw data to help the program learn better. Sometimes, this includes creating new features using expert knowledge and creativity.

Example: using the length and width of a leaf to help identify a plant species

5. Model Evaluation

Checking how well a machine learning model is doing is important to see if it's working well. Measurements like accuracy, precision, recall, F1-score, and mean squared error are often used to check a model's performance.


⏲️ 24 seconds left, you go good!

lets also cover Deep Learning!


Deep Learning Ideas

1. Neural Networks (NN)

This is really abstract. You may imagine an NN as brain-neurons ordered in columns and pinging each other from left to right with different force (via connections) — the force with which a neuron (in each layer) is pinged defines the next ping and the so the end result itself.

Algorithms and problems to solve: recognizing objects in images (feedforward neural networks), translating languages (radial basis function networks)

2. Backpropagation

When you make a mistake, you learn from it and try not to repeat it. Backpropagation is a way for a program to do the same thing.

It helps the program understand where it went wrong and get better at finding the right answers.

3. Convolutional Neural Networks (CNNs)

CNNs are special neural networks that can understand grid-like data, like pictures.

They have layers that help them learn patterns and recognize parts of the picture, such as lines and shapes.

Algorithms and problems to solve: detecting faces in photos (LeNet-5), identifying different types of animals in images (AlexNet, VGG)

4. Recurrent Neural Networks (RNNs)

RNNs are designed to work with data that comes in a sequence, like a series of numbers or words.

They can remember previous inputs and use that information to make better decisions.

Algorithms and problems to solve: predicting stock prices (Long Short-Term Memory, LSTM), generating text based on a given style (Gated Recurrent Units, GRU)

5. Transfer Learning

This is when a program model that has already learned a lot is fine-tuned to work on a new task with limited data.

It helps the program learn faster and perform better because it already knows useful things from its previous learning.

Example: using a model trained on many dog breeds (like ResNet) to recognize specific types of cats

6. Regularization Techniques

These techniques help the program avoid learning too much from the data, which can cause problems like overfitting

STOP 🏁

finished in: 0 min, 52 sec

dominic toretto drives fast as usual

this is basically it, now go and train your agents Kelvin googling implementations for algorithms and problems you want to solve!


to start using a tool, you start with a problem the tool solves and learn what Concept(s) are used to solve the problem

dont keep implementation in your head like "how to do something with a tool", it's complicated and everything won't fit in your head anyway

learn ideas, google implementations

after 5~10 times of googling an implementation you will memorize it, then the technology becomes obsolete and you forget it, its ok, it happens all the time

ideas are hard to forget, they fit in your memory for a long time

bye and see you in the next drag race

gosling-drive-walking.gif


wait actually

think about following the twit bird if you want to see education systems to be practice-first and beneficial <3

anyway you may follow twitter if you just liked the text and want more or you are addictive to fun on social networks

or dont follow anybody and dont listen to anyone! make your own way!

i actually want you to follow my twits, it was just a sale

Check my "Learn REACT in 43 seconds"

Top comments (19)

Collapse
 
dmitry_i_57e465c3faaac74f profile image
Dmitry i

I have to say, your article on machine learning and deep learning was quite the machine-learning experience for me! As someone who is new to these technologies, I appreciated your emphasis on starting with the big picture and problem-solving before diving into the nitty-gritty. And I have to admit, the humorous tone made it feel like I was being entertained as well as educated. Thanks for making learning about AI so fun! Now, if only I can get my neural network to tell me some good jokes like you...

Collapse
 
disukharev profile image
Dima Sukharev

thank you man ;)

Collapse
 
zyabxwcd profile image
Akash • Edited

Nailed It.
Though a little more information like how and why they are different in contrast to each other would better cement the idea. For example, I was left thinking why the structure of the data is important, like for CNN and RNN. Why are they separate just because they work on different structured data. There is no new idea in that. One of them mentioned that it learns patterns, but I thought isn't that what all of them are doing? They are all learning patterns from the data we are training them on. The other one said that it can remembers previous inputs, I couldn't grasp and visualise sort of what it actually meant. Isn't this also what others are doing? The final model can be said to have remembered the previous inputs (as in the input it was trained on) cause it affected how its producing/predicting results. All the inputs that came before affected its decision making.
Tbh, I do have some high level familiarity with ML so I know there are a lot of nuances and maths going on in the background, beneath the abstract ideas but I let you know what comes to mind in general and I think tackling that question is pretty 'clicking' for a new learner.

Collapse
 
disukharev profile image
Dima Sukharev

great thoughts! will do RL on that in the next one ;)

Collapse
 
timofey_klykov_7aebef06c3 profile image
Timofey Klykov

Hey, I just started learning about AI in school and your post helped me a lot! I was kind of confused about the difference between machine learning and deep learning, but now it's much clearer. Thanks for making it so easy to understand!
Baounya

Collapse
 
disukharev profile image
Dima Sukharev

great that you liked it! thanks

Collapse
 
serjobas profile image
Sergey Bunas

This post is a gem! You've made machine learning and deep learning sound much more approachable. Thanks man!

Collapse
 
disukharev profile image
Dima Sukharev

love to know that, bro ;)

Collapse
 
serjobas profile image
Sergey Bunas • Edited

I've teached my ML and it said

"U, me, hello, world, Manner turns the world
In a double cap, my zany, and the other I keep my bean
Uh, I, wait, I'm swervin' shit alone"

Am I wrong for this?

Collapse
 
saeedulkudry profile image
Saeed Adam

Amazing!!!

Collapse
 
disukharev profile image
Dima Sukharev

;)

Collapse
 
kubulashvili profile image
Aleksey Kubulashvili

wow, that's a great way of thinking

Collapse
 
disukharev profile image
Dima Sukharev

it's actually applicable to anything

Collapse
 
exxxcuseme profile image
Daniil Lomovtsev

Learning concepts is a good model to start learning a new technology, but pratice makes deffirence anyway!

Collapse
 
disukharev profile image
Dima Sukharev

agree here

Collapse
 
__c33c7860d4 profile image
Роман Чернышов

Love this! You really managed to simplify some complex ideas.

Collapse
 
disukharev profile image
Dima Sukharev

i love you enjoyed it ;)

Collapse
 
__c33c7860d4 profile image
Роман Чернышов

I'm your fan, bro.

Collapse
 
disukharev profile image
Dima Sukharev