Introduction to

AI/ML/DL & THEIR APPLICATIONS

Human face modeling in real time using single camera of iPhone or computer:

2 examples of AI/ML using Google's TensorFlow and Posenet

ARTIFICIAL INTELLIGENCE, MACHINE & DEEP LEARNING

Artificial Intelligence (AI)

Machine Learning (ML)

Deep Learning (DL)

A technique for implementing ML such as Deep Neural Networks (DNNs) where the code structures are arranged in layers that loosely mimic the human brain, learning patterns of patterns.

An approach to achieve AI through systems that can learn from experience to find patterns in a set of data. "ML is the field of study that gives computers the ability to learn without being explicitly programmed.” - Arthur Samuel, 1959.

"Human intelligence exhibited by machines” - a broad term for getting computers to perform human tasks

Object recognition
Speech recognition
Natural Language Processing
Creative
Prediction
Translation
Restoration
...

The beauty of ML is that it learns by itself from the data passed to it.

ARTIFICIAL INTELLIGENCE, MACHINE & DEEP LEARNING

AlphaFold

Imagined by a GAN (generative adversarial network)

StyleGAN2 (Dec 2019) - Karras et al. and Nvidia

ARTIFICIAL INTELLIGENCE, MACHINE & DEEP LEARNING

TRADITIONAL PROGRAMMING vs. MACHINE LEARNING

ML systems are programmed of course, but the way to achieve their functionality is not given by the programmer. ML programming is about creating algorithms that learn complex functions from data and make predictions on it. These systems can be reused to recognize other objects with new data using the same code.

if email contains V!agra

	then mark as_spam;
    
if email contains ...

if email contains ...

try to classify some emails;
    
change self to reduce errors;

repeat;

MACHINE LEARNING PRINCIPLE

We prepare a big dataset of instances of the problem we want to solve.

The ML system uses the dataset to train itself and create a model of problem we want to solve.

Then the model can be used to predict the answer to new problem instances.

MACHINE LEARNING TYPES

There are two big categories depending on the dataset we provide (and different results we can get):

Supervised learning: we go over the dataset and mark ourselves the right answer.

Unsupervised learning: we give the dataset without hints and let the system figure out patterns.

DEEP LEARNING

Neural Networks (NNs) is a way to make ML system flexible by linking together many simple functions. This is inspired by the neurons that are interconnected and trigger each other.

Deep Neural Networks is NNs with many hidden layers. Each layer will work as if the previous hidden layer is the input it is trying to learn. This is where the name DL comes from as we use the ML paradigm.

DEEP LEARNING WITH TENSORFLOW

DEEP LEARNING WITH CONVOLUTIONAL NNs (CNNs)

DEEP LEARNING WITH REINFORCEMENT LEARNING

Learning by trial-and-error through reward or punishment:

The program learns by playing the game millions of times. We reward the program when it makes a good move. This strengthens the connections to make moves like it did. When it loses we give no reward (or negative reward).

Overtime it learns to maximize reward without the human explicitly telling the rules. It can lead to better than human performance when it finds plays that no one ever thought of doing before ...

PHYSICS-INFORMED NEURAL NETWORK (PINN)

PINN

PHYSICS

Applications to Geoscience

TIMELINE - 70 YEARS OF ML IN GEOSCIENCE

MACHINE LEARNING BASED LOG PREDICTION (TGS)

DEEP LEARNING BASED SALT VMB (TGS)

MULTI-TASK DEEP LEARNING

DEEP LEARNING FOR FAULT & HORIZON INTERPRETATION

F3 example - offshore NL

Although trained by only synthetic datasets, the CNN model works well in predicting faults in field datasets that are acquired at totally different configurations.