Scribd
Upload
25 views12 pages

Master in Artificial Intelligence: Enrichmentors Growing Through Excellence Over 40 Years To Become Best in Management

This document outlines the key components of algorithm selection and development for aspiring AI engineers, focusing on data understanding, algorithm research, and model evaluation. It emphasizes the importance of preprocessing data, selecting suitable algorithms, and using appropriate evaluation techniques like cross-validation and hyperparameter tuning. The content aims to equip learners with the necessary skills to effectively choose and develop algorithms in the field of artificial intelligence.

Uploaded by

andri.wiratmono
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
25 views12 pages

Master in Artificial Intelligence: Enrichmentors Growing Through Excellence Over 40 Years To Become Best in Management

This document outlines the key components of algorithm selection and development for aspiring AI engineers, focusing on data understanding, algorithm research, and model evaluation. It emphasizes the importance of preprocessing data, selecting suitable algorithms, and using appropriate evaluation techniques like cross-validation and hyperparameter tuning. The content aims to equip learners with the necessary skills to effectively choose and develop algorithms in the field of artificial intelligence.

Uploaded by

andri.wiratmono
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

Master in Artificial Intelligence

Algorithm Selection & Development VII


Enrichmentors Growing through Excellence over 40 years to become Best in Management
Purpose
The purpose of the section is to help you learn how
to research, select, and develop appropriate
algorithms to become a Successful Artificial
Intelligence (AI) Engineer

At the end of this lecture, you will learn the


following
• Data Understanding and Preparation
• Researching Algorithms and Architectures
• Model Selection and Evaluation

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Data Understanding and Preparation
Explore and analyze • Characteristics
the available data • Distributions
to understand • Quality

• Missing values
Preprocess the data • Outliers
by handling • Feature scaling

• Training
Split the data into • Validation
• Testing sets for model evaluation

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Researching Algorithms and Architectures

Literature
review
For the
Machine
learning
Deep learning
architectures
problem
research algorithms
domain

Conduct Identify relevant Suitable

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Researching Algorithms and Architectures
Traditional machine learning
algorithms
Linear regression

Decision trees

Support vector machines

Deep learning architectures


Convolutional neural networks

Recurrent neural networks

Transformers

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Researching Algorithms and Architectures

Strengths, weaknesses, and


applicability

Different algorithms and


architectures

Problem requirements,
data characteristics, and
computational resources

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Model Selection and Evaluation

Candidates

Machine learning
algorithms or Deep
learning
architectures

Suitability for the problem and


data

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Model Selection and Evaluation

Using
appropriate
Train and evaluate evaluation
metrics and
validation
techniques

Cross-validation,
Multiple models hyperparameter
tuning

Enrichmentors Growing through Excellence over 40 years to become Best in Management


What is Cross-validation, hyperparameter tuning

Using
appropriate
Train and evaluate evaluation
metrics and
validation
techniques

Cross-validation,
Multiple models hyperparameter
tuning

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Cross-validation
Resampling technique

Involves partitioning the available data

Trained and evaluated

Process repeated multiple times

Performance metrics averaged

K-fold cross-validation, stratified k-fold cross-validation, and leave-one-out cross-validation.

Helps to mitigate issues such as overfitting

Enrichmentors Growing through Excellence over 40 years to become Best in Management


What is next?

Enrichmentors Growing through Excellence over 40 years to become Best in Management


Master in Artificial Intelligence

Algorithm Selection & Development VII


Enrichmentors Growing through Excellence over 40 years to become Best in Management

You might also like