Getting Started with AI

Author(s): Jeff Holmes MS MSCS

Originally published on Towards AI.

How to get started with an AI project

Background

Here I am assuming that you have read my previous article on How to Learn AI. As a reminder, I highly recommend that you refer to more than one resource (other than documentation) when learning ML, preferably a textbook geared toward your learning level (beginner/intermediate / advanced).

What is AI Engineering

AI Engineering is a new discipline focused on developing tools, systems, and processes to enable the application of artificial intelligence in real-world contexts [1]. In a nutshell, AI Engineering is the application of software engineering best practices to the field of AI.

AI vs. ML vs. DL

Artificial intelligence (AI) is the study of ways to build intelligent programs and machines that can creatively solve problems, which has always been considered a human prerogative.

Machine learning (ML) is a subset of AI that provides computer systems the ability to automatically learn and improve from experience without being explicitly programmed. In ML, there are a variety of algorithms that can help solve problems.

Deep learning (DL) is a subset of machine learning that uses neural networks which have a structure similar to the human neural system.

There is often confusion between the terms artificial intelligence and machine learning, which is discussed in The AI Process.

In graduate school, a course in AI will usually have a quick review of the core ML concepts (covered in a previous course) and then cover searching algorithms, game theory, Bayesian Networks, Markov Decision Processes (MDP), reinforcement learning, and more.

MIT Overview of AI and ML

Project Definition

The first step in AI projects is to define the problem.

In a few sentences, describe the following:

• What is the goal? Classify, predict, detect, translate, etc.
• Describe the dataset, including the input features and target feature(s).
• Include summary statistics of the data, including counts of any discrete or categorical features and the target feature.
• Be willing to share the entire dataset. If you cannot share the dataset then you should avoid using open discussion forums such as Slack or Discord.

How to Choose a Model

Every new AI engineer finds that they need to decide what model to use for a problem.

There are many models to choose, but there are usually only slight alterations needed to change a regression model into a classification model and vice versa.

The AI Process describes the AI engineering process for solving AI problems that is applicable to almost any AI problem.

Understanding Algorithms

You need to know what algorithms are available for a given problem, how they work, and how to get the most out of them. However, this does not mean you need to code the algorithms from scratch (roll your own). It is more important to know how to use robust software libraries and frameworks such as scikit-learn and TensorFlow to solve real-world problems.

Any competent software engineer can implement any algorithm.

Even if you are an experienced AI/ML engineer, you should know the performance of simpler models on your dataset/problem.

In fact, there are many concepts that are more important than knowing how AI algorithms work or how to implement them (see Core Concepts in How to Learn AI)

Additional Resources

I have a GitHub repo with lots of notes and links to AI/ML articles on various topics LearnAI. I also have a medium article on AI Learning Resources.

References

[1] Artificial Intelligence Engineering

[2] J. Brownlee, “ Applied Machine Learning Process,” Machine Learning Mastery, Feb. 12, 2014.

[3] J. Brownlee, “ How to Evaluate Machine Learning Algorithms,” Machine Learning Mastery, Aug. 16, 2020.

[4] Y. Hosni, “ Brief Guide for Machine Learning Model Selection,” MLearning.ai, Dec. 4, 2021.

[5] M. LeGro, “ Interpreting Confusing Multiple Linear Regression Results,” Towards Data Science, Sep. 12, 2021.

[6] E. Alpaydin, Introduction to Machine Learning, 3rd ed., MIT Press, ISBN: 978–0262028189, 2014.

[7] S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, 4th ed. Upper Saddle River, NJ: Prentice Hall, ISBN: 978–0–13–604259–4, 2021.

[8] P. Bourque and R. E. Fairley, Guide to the Software Engineering Body of Knowledge, v. 3, IEEE, 2014.

S. Raschka. and V. Mirjalili, Python Machine Learning, 2nd ed. Packt, ISBN: 978–1787125933, 2017.

W. McKinney, Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython, 2nd ed., O’Reilly Media, ISBN: 978–1491957660, 2017.

S. Bird, E. Klein, and E. Loper. Natural Language Processing with Python — Analyzing Text with the Natural Language Toolkit.

D. Jurafsky and J. H. Martin. 2008 (2nd edition). Speech and Language Processing. Prentice Hall. ISBN: 978–0131873216.

Y. Hosni, “ End-to-End Machine Learning Workflow (Part 1),” MLearning.ai, Dec. 11, 2021.

Y. Hosni, “ End-to-End Machine Learning Workflow (Part 2),” MLearning.ai, Dec. 25, 2021.

I. Kilic, “ Data Science Terminology — AI / ML / DL,” Medium, Dec. 31, 2021.

C. Gulmez, “ How to Become an AI Developer,” Medium, Jan. 1, 2022.

D. Gong, “ Practical Guide to Linear Regression,” Towards Data Science, Sept. 18, 2021.

Y. Sandeepa, “ Regression for Classification,” Towards Data Science, Sept. 13, 2021.

A. Iglesias Moreno, “ End-to-end machine learning project: Telco customer churn,” Towards Data Science, Nov. 23, 2021.

M. Di Pietro, “ Deep Learning with Python: Neural Networks (complete tutorial),” Towards Data Science, Dec. 17, 2021.

D. Poulopoulos, “ How to start contributing to open-source projects,” Towards Data Science, Oct. 30, 2021.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI