Artificial Intelligence

ACTL3143 & ACTL5111 Deep Learning for Actuaries

Patrick Laub

Artificial Intelligence

Lecture Outline

Artificial Intelligence
Deep Learning Successes (Images)
Deep Learning Successes (Text)
Classifying Machine Learning Tasks
Neural Networks

Different goals of AI

Artificial intelligence describes an agent which is capable of:

Thinking humanly	Thinking rationally
Acting humanly	Acting rationally

AI eventually become dominated by one approach, called machine learning, which itself is now dominated by deep learning (neural networks).

There are AI algorithms for simple tasks that don’t use machine learning though.

Shakey the Robot (~1966 – 1972)

Route-finding I

At its core, a pathfinding method searches a graph by starting at one vertex and exploring adjacent nodes until the destination node is reached, generally with the intent of finding the cheapest route. Although graph searching methods such as a breadth-first search would find a route if given enough time, other methods, which “explore” the graph, would tend to reach the destination sooner. An analogy would be a person walking across a room; rather than examining every possible route in advance, the person would generally walk in the direction of the destination and only deviate from the path to avoid an obstruction, and make deviations as minor as possible. (Source: Wikipedia)

Route-finding II

Tunes of the Kingdom: Evolving Physics and Sounds for ‘The Legend of Zelda: Tears of the Kingdom’, GDC 2024

Evaluating a chess game I

Who’s winning this game?

	5 × 1 = 5
	0 × 3 = 0
	2 × 3 = 6
	2 × 5 = 10
	0 × 9 = 0
	1 × 0 = 0
White	21

Evaluating a chess game II

Just add up the pieces for each player.

	6 × 1 = 6
	1 × 3 = 3
	1 × 3 = 3
	2 × 5 = 10
	0 × 9 = 0
	1 × 0 = 0
Black	22

Overall: 21 − 22 = −1.

The minimax algorithm

Chess

Deep Blue (1997)

Machine Learning

Tried making a computer smart, too hard!

Make a computer that can learn to be smart.

The Venn diagram of Artificial Intelligence, Machine Learning, and Deep Learning.

Definition

“[Machine Learning is the] field of study that gives computers the ability to learn without being explicitly programmed” Arthur Samuel (1959)

Deep Learning Successes (Images)

Lecture Outline

Artificial Intelligence
Deep Learning Successes (Images)
Deep Learning Successes (Text)
Classifying Machine Learning Tasks
Neural Networks

Image Classification I

What is this?

Options:

punching bag
goblet
red wine
hourglass
balloon

Note

Hover over the options to see AI’s prediction (i.e. the probability of the photo being in that category).

Image Classification II

What is this?

Options:

sea urchin
porcupine
echidna
platypus
quill

Image Classification III

What is this?

Options:

dingo
malinois
German shepherd
muzzle
kelpie

ImageNet Challenge

ImageNet and the ImageNet Large Scale Visual Recognition Challenge (ILSVRC); originally 1,000 synsets.

AlexNet — a neural network developed by Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton — won the ILSVRC 2012 challenge convincingly.

How were the images labelled?

“Two years later, the first version of ImageNet was released with 12 million images structured and labeled in line with the WordNet ontology. If one person had annotated one image/minute and did nothing else in those two years (including sleeping or eating), it would have taken 22 years and 10 months.

To do this in under two years, Li turned to Amazon Mechanical Turk, a crowdsourcing platform where anyone can hire people from around the globe to perform tasks cost-effectively.”

Needed a graphics card

A graphics processing unit (GPU)

“4.2. Training on multiple GPUs A single GTX 580 GPU has only 3GB of memory, which limits the maximum size of the networks that can be trained on it. It turns out that 1.2 million training examples are enough to train networks which are too big to fit on one GPU. Therefore we spread the net across two GPUs.”

Lee Sedol plays AlphaGo (2016)

Deep Blue was a win for AI, AlphaGo a win for ML.

Lee Sedol playing AlphaGo AI

I highly recommend this documentary about the event.

Generative Adversarial Networks (2014)

https://thispersondoesnotexist.com/

Diffusion models

Painting of avocado skating while wearing a hoodie

A surrealist painting of an alpaca studying for an exam

Dall-E 2 (2022) vs Dall-E 3 (2023)

Same prompt: “A beautiful calm photorealistic view of an waterside metropolis that has been neglected for hundreds of years and is overgrown with nature”

Deep Learning Successes (Text)

Lecture Outline

Artificial Intelligence
Deep Learning Successes (Images)
Deep Learning Successes (Text)
Classifying Machine Learning Tasks
Neural Networks

GPT

AI predictions in the classification demo were from GPT code.

Homework Get ChatGPT to:

generate images
translate code
explain code
run code
analyse a dataset
critique code
critique writing
voice chat with you

Compare to Copilot.

Code generation (GitHub Copilot)

Students get Copilot for free

Use a free trial then sign up for free education account

A student post from last year:

I strongly recommend taking a photo holding up your Academic Statement to your phone’s front facing camera when getting verified for the student account on GitHub. No other method of taking/uploading photo proofs worked for me. Furthermore, I had to make sure the name on the statement matched my profile exactly and also had to put in a bio.

Good luck with this potentially annoying process!

Homework It’s a slow process, so get this going early.

Programmers are increasingly using AI

Question: What is your experience with the following AI tools?

Classifying Machine Learning Tasks

Lecture Outline

Artificial Intelligence
Deep Learning Successes (Images)
Deep Learning Successes (Text)
Classifying Machine Learning Tasks
Neural Networks

A taxonomy of problems

Machine learning categories in ACTL3142.

New ones:

Reinforcement learning
Semi-supervised learning
Active learning

Supervised learning

The main focus of this course.

Regression

Given policy \hookrightarrow predict the rate of claims.
Given policy \hookrightarrow predict claim severity.
Given a reserving triangle \hookrightarrow predict future claims.

Classification

Given a claim \hookrightarrow classify as fraudulent or not.
Given a customer \hookrightarrow predict customer retention patterns.

Supervised learning: mathematically

A recipe for supervised learning.

Self-supervised learning

Data which ‘labels itself’. Example: language model.

‘Autoregressive’ (e.g. GPT) versus ‘masked’ model (e.g. BERT).

Example: image inpainting

Other examples: image super-resolution, denoising images.

Example: Deoldify images #1

A deoldified version of the famous “Migrant Mother” photograph.

Example: Deoldify images #2

A deoldified Golden Gate Bridge under construction.

Neural Networks

Lecture Outline

Artificial Intelligence
Deep Learning Successes (Images)
Deep Learning Successes (Text)
Classifying Machine Learning Tasks
Neural Networks

How do real neurons work?

A neuron ‘firing’

An artificial neuron

A neuron in a neural network with a ReLU activation.

One neuron

\begin{aligned} z~=~&x_1 \times w_1 + \\ &x_2 \times w_2 + \\ &x_3 \times w_3 . \end{aligned}

a = \begin{cases} z & \text{if } z > 0 \\ 0 & \text{if } z \leq 0 \end{cases}

Here, x_1, x_2, x_3 is just some fixed data.

A neuron in a neural network with a ReLU activation.

The weights w_1, w_2, w_3 should be ‘learned’.

One neuron with bias

\begin{aligned} z~=~&x_1 \times w_1 + \\ &x_2 \times w_2 + \\ &x_3 \times w_3 + b . \end{aligned}

a = \begin{cases} z & \text{if } z > 0 \\ 0 & \text{if } z \leq 0 \end{cases}

The weights w_1, w_2, w_3 and bias b should be ‘learned’.

Bias = -4
0
4

A basic neural network

A basic fully-connected/dense network.

Step-function activation

Perceptrons

Brains and computers are binary, so make a perceptron with binary data. Seemed reasonable, impossible to train.

Modern neural network

Replace binary state with continuous state. Still rather slow to train.

Note

It’s a neural network made of neurons, not a “neuron network”.

Try different activation functions

Flexible

One can show that an MLP is a universal approximator, meaning it can model any suitably smooth function, given enough hidden units, to any desired level of accuracy (Hornik 1991). One can either make the model be “wide” or “deep”; the latter has some advantages…

Feature engineering

Doesn’t mean deep learning is always the best option!

Quiz

In this ANN, how many of the following are there:

features,
targets,
weights,
biases, and
parameters?

What is the depth?

Package Versions

from watermark import watermark
print(watermark(python=True, packages="keras,matplotlib,numpy,pandas,seaborn,scipy,torch,tensorflow,tf_keras"))

Python implementation: CPython
Python version       : 3.11.9
IPython version      : 8.24.0

keras     : 3.3.3
matplotlib: 3.9.0
numpy     : 1.26.4
pandas    : 2.2.2
seaborn   : 0.13.2
scipy     : 1.11.0
torch     : 2.3.1
tensorflow: 2.16.1
tf_keras  : 2.16.0

Glossary

activations, activation function
artificial neural network
biases (in neurons)
classification problem
deep network, network depth
dense or fully-connected layer
feed-forward neural network
labelled/unlabelled data
machine learning

minimax algorithm
neural network architecture
perceptron
ReLU
representation learning
sigmoid activation function
targets
weights (in a neuron)