Learn AI

through games, not lectures

Learn reinforcement learning in 4 weeks. Every week, build an AI which battles to be crowned champion of the cohort in a live competition.

Next cohort starts 11th July.

Can't make this cohort?Stay updated via our mailing list

Taught by a team from:

Learn, Build & Compete

in live team games

Online courses are rarely fun. It’s easy to lose motivation and give up.

Delta Academy combines live, competitive team coding games with interactive tutorials that teach machine learning to a peer group.

Play to win

& learn Machine Learning

In each Delta Academy competition, work as a team to build a game AI and compete against others.

Get up to Speed

Get introduced to new concepts in code through short interactive tutorials that prepare you for the competition at the end of the week.

Team Up

Software is built by teams, not individuals. That's why we encourage collaborating in pairs in competitions. Form your dream-team: bring a friend, or make new ones!

Strive for Victory

Get competitive. Unlike dull online tutorials, where there’s nothing on the line, find yourself ultra-motivated as you strive for victory!

Reinforcement Learning:

Learn from Experience

Deep RL is one of the most impactful recent developments in AI. It's responsible for many of the most impressive breakthroughs, including beating the World Champion in Go (AlphaGo). That's why it's the focus of the 4-week Delta Academy course.

Cutting-Edge Code

Learn PyTorch, the machine learning framework built by Facebook AI Research used by researchers and practitioners in industry.

Stuck? Here to help!

Experts are always on hand to immediately answer questions and help you out.

What students say

I get to meet some amazing people and get to see how they would solve a different problem. I love it!

The fact that there's a new competition every week compels you to get better fast.

I like it very much, since it encourages team work, provides help when needed. It's really focused on teaching as well as solving problems which is unlike Leetcode or something like Codeacademy. On a whole different spectrum.

Experience. And a shared experience at that. It is not only how I solve a problem, but also how my partner solves it, and how other participants solve it.

A Unique Master's Level Syllabus

Intro to Deep Reinforcement Learning

A unique 4-week course, taught by experts from Oxford, Cambridge and DeepMind. Learn the theory and immediately implement Reinforcement Learning algorithms in PyTorch in weekly AI-building competitions.
Join a cohort of peers to collaborate with and compete against. (exp. time commitment: 10 hrs per week)

Includes

  • 12 Expert-Crafted Interactive Tutorials

  • 4 RL Competitions to Play

  • Cohort of Peers to Learn and Compete with

  • Expert Teaching & Code Review

Pre-requisites

  • Basic Python - Loops, Functions & Data Types

  • Basic Probability

Starts July 11

th

Per week for 4 weeks.

$14.99USD
Cancel anytime

Course Organisation

4 Weeks from July 11th to August 6th

Fully remote. Learn from anywhere.

A new way to learn tech skills

Expert-Crafted Tutorials

Each week starts with 3 tutorials explaining new concepts, each with problems to solve to ensure you can put what you're learning into practice.

Compete Every Week

Apply what you've learned each week in the competition. The challenge is released 24 hours before the deadline. We recommend you spend 2 hours on it.

Live Discussion & Competition

Participate in a live discussion of how each solution works, then watch the AI's compete! Lastly, learn about how to build an optimal solution & see the code from the experts.

Meet your Instructors

Dr. James Rowland

University of Oxford

James completed his PhD from the University of Oxford in Neural Computation, studying the paths information takes through the mammalian brain. He's since worked as a Data Scientist at early-stage startups, implementing machine learning models.

Henry Pulver

Cambridge University, Five AI

While at Cambridge University studying for a Masters' in Machine Learning, Henry wrote his thesis on Reinforcement Learning. He then published papers as a machine learning researcher at the UK's largest autonomous driving startup, Five AI.

Dr. Matthew Phillips

DeepMind, University College London

Matt's research at DeepMind was in Multi-agent Reinforcement Learning - how autonomous agents trade-off collaboration and competition. His PhD from University College London is in the Neuroscience of learning and memory, so deeply understands teaching.

Course Syllabus

Week 1

Reinforcement Learning Fundamentals

What kinds of problems can Reinforcement Learning solve? How does it work? What is the trade-off between exploration and exploitation? The first week answers these questions and more, plus you'll build your first Reinforcement Learning algorithm to play a well-known game and compete amongst the cohort!

Week 2

Learning from Experience

Learn about Monte Carlo and Temporal Difference Learning, two of the key Reinforcement Learning algorithms. You'll code up both of these algorithms and understand how they work, their benefits & drawbacks and how they fit into the overall picture of Reinforcement Learning.

Week 3

Deep Neural Networks with PyTorch

Onto neural networks - from first principles, learn how to design and build Deep Neural Networks with PyTorch. Then apply them using the RL algorithms you've used thus far to build your first Deep Reinforcement Learning agents!

Week 4

Deep Q-Networks

DQN finds approximate solutions to much more complex RL problems than those we've been able to tackle thus far. Plus, it can do so without any prior information about the environment it acts in. We'll learn how to implement DQN and insider information on efficient training.

Interested in joining the cohort?

Join the 4-Week Intro to Reinforcement Learning cohort starting 13th June while there are still spaces!

Join Cohort