A link list of machine learning resources, curated by the ML team at SC5.
- AI, Deep Learning, and Machine Learning: A Primer - A 45 minute video on the history behind AI and machine learning. A must-watch for anyone interested in the field.
- What Artificial Intelligence Can and Can’t Do Right Now - A down-to-earth explanation of what's currently possible (and not possible). Focuses mostly on supervised learning.
- Ten Myths About Machine Learning - Everyone should read this before blinding trusting everything reported by the media. By Pedro Domingos.
- Meet the People Who Train the Robots (to Do Their Own Jobs) - Includes interviews with a travel agent, a robotics expert, the CEO of a legal services startup, a customer service representative, and a software engineer from Waymo.
- Bayesian machine learning - A great high-level overview of what machine learning is from a Bayesian viewpoint. An extremely useful primer if you already have some knowledge of "traditional" machine learning methods like logistic regression.
- An Introduction to Deep Learning - A good, plain English overview of deep learning.
- Deep Learning - The book on deep neural networks. Covers all of the maths, intuition and lots of different architectures. If you aren't comfortable with matrices and linear algebra, it'd be a good idea to read up on those first before picking this up.
- Deep Learning Simplified - A video series that gives a non-technical explanation of deep learning. Highly recommended for beginners.
- A Tutorial on Deep Learning (PDF) - A great 2-part series by Quoc V Le on Deep Learning. Covers the basics and some more advanced algorithms such as convolutional neural networks, autoencoders and recurrent neural networks. Part 2 can be found here.
- Nuts and Bolts of Applying Deep Learning - A one-hour workshop video in which the legendary Andrew Ng explains the practical issues surrounding Deep Learning, along with tips on how to solve them.
- Reinforcement Learning by David Silver from DeepMind - A great series of videos (15 hours) on reinforcement learning by David Silver. Technical, but very much worth it. A great place to start if you're a beginner. Requires some math knowledge, but don't let that deter you. The slides for the lectures can be found here.
- Demystifying Deep Reinforcement Learning - A good tutorial that focuses on Q-learning using deep neural networks.
- Reinforcement Learning: An Introduction - A comprehensive book detailing lots of different reinforcement learning algorithms and techniques.
- Deep Reinforcement Learning: Pong from Pixels - A great, in-depth piece on the policy-based Policy Gradient reinforcement learning algorithm.
- Deep Q Learning with Keras and Gym - An overview of the Deep Q Network (DQN) algorithm for reinforcement learning. Includes full Keras code.
- How to measure importance of inputs? - What not to do when interpreting how different inputs affect the output of a machine learning model.
- Embed, encode, attend, predict: The new deep learning formula for state-of-the-art NLP models - Does what it says on the tin. Rather complex, but a good read if you have intermediate knowledge of neural networks and NLP concepts.
- An Interactive Tutorial on Numerical Optimization - A great interactive visualisation of different types of numerical optimisation algorithms, including Gradient Descent. Helps you get a feel for how an optimisation algorithm actually behaves in motion.
- The Neural Network Zoo - A great blog post that visualises different types of neural networks, along with well-written descriptions explaining why you might want to choose one over the other.
Brought to you, with love, by
