Scribd
Upload
4 views

ML Lecture

The document provides a comprehensive overview of machine learning concepts, focusing on applications with satellite data. It covers various learning types such as supervised, unsupervised, and reinforcement learning, along with techniques like neural networks and convolutional layers. Additionally, it discusses the importance of data quality and the challenges faced when applying CNNs to satellite imagery.

Uploaded by

MD. AL MAMUN
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
4 views

ML Lecture

The document provides a comprehensive overview of machine learning concepts, focusing on applications with satellite data. It covers various learning types such as supervised, unsupervised, and reinforcement learning, along with techniques like neural networks and convolutional layers. Additionally, it discusses the importance of data quality and the challenges faced when applying CNNs to satellite imagery.

Uploaded by

MD. AL MAMUN
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 41

Machine Learning with

Satellite Data

A blisteringly-fast crash course

Sean Foley
What is Machine Learning?
Linear Algebra Optimization
● The language used by the other fields ● Make number go up!
● Vectors, matrices, spaces ● Iterative techniques
● Numerical linear algebra: how to go fast! ● Gradient-based methods
● Leverage hardware

Signal Processing
● Signal vs. noise
● Mutual information / entropy Data Science
● Bits! ● Garbage in, garbage out
● 90% of the work is getting good data
● The most important part (in my opinion)
High-Dimensional Probability & Stats.
● If I throw one trillion twenty-sided dice…
● Surprisingly geometrical
● Some intuitive concepts break down in “The unreasonable effectiveness of data” - Peter
higher dimensions! Norvig
Supervised Unsupervised Reinforcement
Learning Learning Learning
Inputs Data
Agent

Model Method ● Observations


Actions
● Rewards

Loss Function Classes / clusters


Targets /
Environment
Predictions
Labels
Supervised Unsupervised Reinforcement
Learning Learning Learning

● K-Means Clustering
● Decision Trees
● Principal Component
● Random Forests
Analysis (PCA)
● Support Vector
● Gaussian Mixture
Machines (SVM)
Models

● Neural Networks + ● Self-Supervised ● Monte Carlo RL


Backpropagation Learning ● Deep Q-Learning
● Asynchronous
Advantage Actor-Critic
Variants
● Semi-supervised learning
● Transfer learning
● Limited supervision
○ Low-shot / few-shot
○ Weakly supervised
Regression vs. Classification
P(x, y)
Supervised
Learning
Inputs

Model

Loss Function
Targets /
Predictions
Labels
Supervised
Learning
Inputs

Model

Loss Function
Targets /
Predictions
Labels
Regress chlorophyll-a

Loss
x Slide credit: Patrick Gray
Classify phytoplankton types
chlorophyll-a

chlorophyll-a
Synechococcus Synechococcus

Prochlorococcus Prochlorococcus

phycoerythrin phycoerythrin

Slide credit: Patrick Gray


Neural Network / Multi-Layer Perceptron (MLP)
Neural Network / MLP
Adding more layers?

Uh oh!!
Activation Functions
● Non-linear
● Historically:
○ Sigmoid
○ Tanh
● Modern:
○ ReLU
○ Leaky ReLU

Et voila…
Loss Function

● Differentiable
● Aligns with evaluation metrics
● e.g.
○ Mean Squared Error (L2 loss) <- regression

○ Cross Entropy Loss <- classification


W1 W2
Backpropagation
Chain rule!

Loss
Function
Gradient Descent

W1 W2
Gradient Descent

● Step size
● Stochastic Gradient Descent (SGD)
● Other Gradient-based methods
○ Adam: most commonly-used
Evaluation Metrics (Classification)
Evaluation Metrics (Regression)

● Mean Absolute Error (MAE) aka L1 Loss


● Mean Squared Error (MSE) aka L2 Loss
Instances, Batches, Epochs

● Instance: one sample


● Batch: bundle of instances
● Batch size!!
○ Stochasticity
○ Convergence speed
● Epochs
● Early stopping
Training, Validation, Testing

● Generalizability!
● Train: parameters
● Validation: hyperparameters
● Testing: nothing!
Training / validation script overview
Over/under fitting, revisited
underfitting overfitting
Graphics Processing Units (GPUs)

Optimized for large matrix operations

High-level code

CUDA

NVIDIA Graphics Card


Satellite
Imagery

Spatial context!

Translational
symmetry!
Convolution

1D, continuous convolution shown here

flip g!
2D discrete convolution
Image Cubes
Height

RGB

Height
OCI

Width

80
~2
Width
A Convolutional Layer: Basics

● Fully-connected in the channel dimension


● Sliding window in the spatial dimensions
A Convolutional Layer: More Details

● bias
● activation function
● differentiable

# parameters:

doesn’t depend on image size!!


A Convolutional Layer: Padding
Receptive Field
An Example CNN

Very cool interactive demo:

https://adamharley.com/nn_vis/cnn/2d.html
Data Augmentation
CNNs + Satellite Data: Caveats

Most CNNs are NOT designed for satellite data.

Especially not for Earth Science data!


● vast scale
● not multi-scale
● # channels
Resources / Credit
● Cheat sheets on Stanford AI / ML courses (many figures in this presentation)
● Figures from pytorch docs
● Amazing animations from this article
● Interactive CNN demo
● Many figures from Wikipedia

You might also like