Scribd
Upload
30 views3 pages

ML - AI Presentation

Uploaded by

beaustroms
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
30 views3 pages

ML - AI Presentation

Uploaded by

beaustroms
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/ 3

I.

Introduction

A. Attention Getter (Hook the audience in with a startling statistic, a fun fact, an
interesting example, etc.)

According to National University, nine out of ten students want to learn more about AI in
school.

B. Significance/interest/topic importance (The reason to listen, why is this topic


important?)

AI/ML are becoming omnipresent, especially with generative AI becoming so


commonplace. It’s important to understand technology in order to make best use of it.

C. Thesis- One sentence summary of the central point of your speech.

“By understanding the field of machine learning, anyone can understand the mechanics
of this topic otherwise relegated to science fiction. ”

D. Preview- Identify what your main points will be.

What is “AI” / different subcategories of it

What is Machine Learning -> brief history of machine learning

How do generative models work -> different methodologies

How does it actually work in practice -> Torch, tensorflow, etc

II. Body

A. AI is a broad umbrella. There are many terms under the AI umbrella, but the most
commonly understood is “Machine Learning”. AI in games for example does not learn,
but is prewritten.

*Machine learning refers to a machine capable of adapting to “fit” data, or in other


words to produce outputs consistent with what it’s given.

B. Machine learning was invented by Alan Turing in the 1950s, but at the time the
name had yet to be coined. In 1957, the perceptron was developed by Frank
Rosenblatt, which was a neural network for binary classification. In 1986 Geoffrey
Hinton and others published a paper on backpropagation. In 1997 Deep Blue, a model
by IBM, beat Garry Kasparov in chess. In 2006 deep learning was coined. To continue
on would be endless, as the time gaps between advancements shrink exponentially.
*Today, generative models like ChatGPT represent a large part of both the market
and ML’s place in collective consciousness.

C. Generative models use a variety of algorithms which are dramatically different from
typical classification/regression problems. Among these, Generative Adversarial
Networks (GANs) and Diffusion models are the two most prevalent.

● Transition to conclusion

III. Conclusion

A. Machine Learning and AI, while complicated, are not magic. Lots of different
methodologies go into the umbrella known as AI, all of which are at the end of the day
math. Similarly, the creation of models is not incredibly abstract, but is immediately
attainable.

B. Many people are completely unaware of the particulars involving machine learning,
which can make operation of tools involving it more difficult and frustrating.

C. If you yourself have an interest, Tensorflow Playground is a website which allows


you to tinker with simple machine learning models in an approachable way.

Bibliography:

https://www.nu.edu/blog/ai-statistics-trends/ - National university, accessed


10/31/2024, unknown

https://www.pewresearch.org/social-trends/2023/07/26/which-u-s-workers-are-mo
re-exposed-to-ai-on-their-jobs/ - Pew Research Center, accessed 10/31/2024, Rakesh
Kokhhar

https://www.hiringlab.org/2024/03/14/march-2024-us-labor-market-update/ -
Indeed Hiring Lab, accessed 10/31/2024, Nick Bunker

You might also like