Scribd
Upload
29 views

Abstract - ILT ML 04 T - M0090969

Uploaded by

Hjkk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
29 views

Abstract - ILT ML 04 T - M0090969

Uploaded by

Hjkk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 21

Main topics in video

1. introduction about ML
2. Inftroduction model machine learning
3. Discustion fairness about machine learning

There ara some prerequisites


-basic python programming
-basic matrix algebra
-exposure to statistics and probability

Applied machine learning to solving problem,


machine learning is mathematical models to solve a complex problem

Machine learning is about algorithm to build somethong to machine that can learning from the
examples that wa are given
Implememtation Machine Learning for home prices prediction

Machine Learning compared with traditional programming


Prediction with machine learning is we feed give examples to the models to find the pattern

Mechine learning learnig from examples


Will learning by exampkes that given

The illustrations how machine learning learns is like a kid point a lion then he says is a cat
A lion is predicted as a cat
And we give examples and give characteristics about line more details that a kid can learn about lion
And when the kid predicts a tiger a s lion until kid knows that a lion truly a lion by the feedbacks that
are we given
Rule-based

Humans develop the logic flow


If/then/else. Rules
One-off exceptions
Limited to human reasoning
Exposes bias of the implementor
Machine Learning
Model trained with data
Emergent patterns in the data
Experiments and refinements in training and models
Limited by training data
Exposes bias of the training data
Highlight
When we want create models you must have a clear definition of what is the problem you want to
tackle
Artificial Inteligence
Machiness that appear intelegent based on the tasks they perform
Basically artificial inteligence is to replcate what human can do, how to human can predict home
prices etc

Machine Learning
A spesific field of AI where a system to find paterns in examples,tpycally using statistics in order to
make predictions
Deep learning
A machine learning approach that breaks a problem down into many pieces. So it can learn more
from the training data

Deep laerning is replicate human brains how neuron in our brains to make a decision

The relation artificial inteligenece , machine learning, deep learning


So machine learning is a part of artificial intekigence and deep learning is part of machine learning
Build system to recommend vidio that are match with user experince or user history video

Using regression to predict prices of house


using machine learning to predict ad that are match wit the targets market

Models Is amthematically way to the patterns and insights that a machine learning system learns
from examples and is to make predictions
Training the models
Features
Information drawn from examples which distinguish one example from another are the features in
machine learning system

Overfitting
When a model confroms too much to its training data and generalize to make predictions about new
data
Classification using labeled data as a guide, detemine which class of category an example belongs to

regression
sequences
Dataet quality
Ehat atributes a dataset would you consider ideal
Larger size/high diversity
Larger size/low diversity
Small size/high diversity
Small size/low diversity
Types of bias
Reporting
Automation
Selection(coverage, non response,sampling)
Group attribution (infroup, out group)
Implicit (confirmation,experiments)

You might also like