Resources
Are you new to the world of bird song and machine learning? We’ve compiled a list of resources to help you get started and demystify the work we do.
Bird song
An introduction to bird song
This article, created by the Cornell Lab of Ornithology, is a great starting point to learn more about bird song and how and why birds sing.
Bird song interactive
Do you prefer something more visual? Check out this interactive on bird song, also developed by the Cornell Lab of Ornithology.
Bird song hero
New to spectrograms? The interactive below will help you understand how spectrograms work and how you can use them to get better at identifying songbirds. You can also look at our spectrograms of roroa, weka, and ruru calls or take a quiz to test your new skills!
Machine learning and deep learning
Neural networks and deep learning
3blue1brown created a great set of resources on neural networks and deep learning. The resources are available both as a set of videos and as a set of articles. Be sure to check the links in the video descriptions if you’d like to go more in depth.
But what is a neural network?
What are the neurons, why are there layers, and what is the math underlying it?
Gradient descent, how neural networks learn
How gradient descent works in the context of neural networks.
What is backpropagation really doing?
What’s actually happening to a neural network as it learns?
Backpropagation calculus
The math of backpropagation, the algorithm behind how neural networks learn.
Learning to see and neural networks demystified
A great set of resources by Welch Labs on machine learning and neural networks.
Learning to see
This series explores the use of machine learning and artificial intelligence through one example from the field of computer vision: using a decision tree to count the number of fingers in an image.
Neural Networks Demystified
This series show how to build and train a complete Artificial Neural Network using python.
More on neural networks
Keen to know more? Check out the resources below.
A gentle introduction to graph neural networks
Neural networks have been adapted to take advantage of the structure and properties of graphs. This article explores the components needed for building a graph neural network – and explains the motivations behind their design choices.
Understanding convolutions on graphs
Understanding the building blocks and design choices of graph neural networks.
Visualizing neural networks
See and view the process and the results of neural networks at work.
Have a question?
We love talking about our kiwi! Read through our list of Questions & Answers, or Get in Touch with your own question.
Our resources page is supported by funding from:
