Machine learning

tags: Artificial Intelligence, Applied maths

Machine learning is about constructing algorithms that can approximate complex functions from observations of input/output pairs. Machine learning is related to Statistics since its goal is to make predictions based on data.

Examples of such functions include:

Image classification
Time-series prediction
Language modeling

Regression

The goal is to approximate a target function \(f\) or signal \(S\).

Links to this note

Adversarial examples
AI and climate change
Alternative learning mechanisms
Artificial Intelligence
Automated theorem proving
Bias-variance tradeoff
Catastrophic forgetting
Computer vision
Continual learning
Data representation
Distillation
Federated learning
Few-shot learning
Gaussian Processes
Generalization in Machine learning
Generative modelling
Homomorphic encryption
Implicit neural representations
Kernel Methods
Learning in dynamical systems
Meta-learning
Neural network training
Neural networks
NLP
Notes on: Generalization over different cellular automata rules learned by a deep feed-forward neural network by Aach, M., Goebbert, J. H., & Jitsev, J. (2021)
Notes on: Next Generation Reservoir Computing by Gauthier, D. J., Bollt, E., Griffith, A., & Barbosa, W. A. S. (2021)
Notes on: One model for the learning of language by Yang, Y., & Piantadosi, S. T. (2022)
Privacy-preserving machine learning
Pytorch
Rademacher complexity
Recurrent neural networks
Reinforcement learning
Reservoir computing
Self-supervised learning
Softmax
Supervised learning
Talk: Artificial Intelligence: A Guide for Thinking Humans
Talk: Differentiation of black-box combinatorial solvers
The Bitter Lesson
Transfer learning
Unsupervised learning
Why programming is a good medium for expressing poorly understood and sloppily-formulated ideas
Zero-shot learning

Last changed 19/04/2022 | authored by Hugo Cisneros

Comments

← Back to Notes