Principles of Data Science
An introduction to data science covering the data lifecycle, statistics, machine learning, deep learning basics, neural networks and ethical considerations in data-driven work.
Search titles, authors, subjects and dates from across the DARE African Knowledge Repository.
10 resources found in Artificial Intelligence & Machine Learning.
Scoped to Artificial Intelligence & Machine Learning · Search all DARE
An introduction to data science covering the data lifecycle, statistics, machine learning, deep learning basics, neural networks and ethical considerations in data-driven work.
A widely used introductory text on neural networks and deep learning covering backpropagation, improving neural-network learning, overfitting, regularisation and deep convolutional networks, with hands-on code examples.
A text on representation learning for NLP: word embeddings, neural language models, pre-trained representations and their use in downstream natural-language-processing tasks.
An introduction to neural networks and classical machine learning with a personal, historically grounded perspective on the shallow/deep distinction and foundational learning algorithms.
Fifteen-module postgraduate course catalogue in AI for science, including Modern Statistics and Machine Learning for Population Health in Africa and Deep Learning for Ecology with African wildlife case studies, offered by AIMS South Africa with Google DeepMind.
Covers geometry, deep multi-view methods, differentiable rendering, generative modeling, representation learning, and vision for embodied agents.
Reviews linear algebra with applications to probability, statistics, optimization, signal processing, and deep learning.
Introductory deep learning course with applications to computer vision, NLP, biology, and TensorFlow practice.
Explores synaptic connectivity, perceptrons, recurrent networks, backpropagation, Hebbian learning, and models of perception and memory.
Practical introduction to deep neural networks, CNNs, transformers, generative LLMs, and text-to-image models.