Skip to content
AI Masters DoJo
ExploreMenu
Book a discovery call

Reinforcement learning

Curiosity-Driven Deep Reinforcement Learning

Explore deep reinforcement learning and research on curiosity-driven exploration in environments without rewards.

A small sample of 30,000+ courses curated from around the world, all available to DoJo members.

Book a discovery callExplore DoJo membership

A free 30-minute conversation with Ted about your goals.

Already a member? Open this course

Course topics

Reinforcement learning

  • A3C foundations
  • Research-paper analysis
  • Curiosity-driven exploration

You observe the behaviour

Find your fit

Is this a useful starting point?

For advanced reinforcement-learning students with experience coding actor-critic agents.

Starting point
Expert
Course length
4 total hours
Lessons
27 lectures

Course by Phil Tabor

Course last updated: 07/2023

Requirements
  • Experience in coding actor critic agents
What is included
  • 4 hours on-demand video
  • Access on mobile
  • Certificate of completion

DoJo membership: A$250 per month per seat, excluding GST. Includes fortnightly group support and a private member community.

Explore the lessons

Course curriculum.

5 sections • 27 lectures • 4h 9m total length

Introduction3 lectures • 12min
  1. What You Will Learn in this Course05:25
  2. How to Succeed in this Course03:44
  3. Required Background, Software, and Hardware03:01
Fundamental Concepts4 lectures • 26min
  1. A Brief Review of Deep Reinforcement Learning and Actor Critic Methods09:45
  2. Code Review of Basic Actor Critic Agent09:17
  3. A Crash Course in Asynchronous Advantage Actor Critic Methods05:41
  4. Our Code Structure01:42
Paper Analysis: Asynchronous Methods for Deep Reinforcement Learning12 lectures • 2hr 5min
  1. How to Read and Implement Research Papers06:54
  2. A3C Paper: Abstract and Introduction04:15
  3. Crash Course in Parallel Processing in Python11:39
  4. A3C Paper: Related Work, Reinforcement Learning Background05:09
  5. A3C Paper: The Asynchronous Reinforcement Learning Framework07:59
  6. Coding our Actor Critic Network11:50
  7. Learning with Generalized Advantage Estimation13:59
  8. Coding a Minimalist Replay Memory03:27
  9. Coding the Shared Adam Optimizer03:28
  10. A3C Paper: Experiments and Discussion10:00
  11. How to Modify the Open AI Gym Atari Environments19:27
  12. Coding Our Main Loop and Evaluating Our Agent27:10
Paper Analysis: Curiosity Driven Exploration by Self Supervised Prediction6 lectures • 1hr 2min
  1. Paper Overview01:17
  2. ICM Paper: Abstract and Introduction06:45
  3. ICM Paper: Curiosity Driven Exploration07:38
  4. Experimental Setup and Coding Our ICM Module27:54
  5. ICM Paper: Experiments, Related Work, and Discussion11:17
  6. Setting Up the Mini World and Training Our ICM Agent06:51
Appendix2 lectures • 24min
  1. Setting Up Our Virtual Environment for the New Open AI Gym09:09
  2. Making Our Agents Compliant with the New Gym Interface14:31

From learning to practice

Try this next.

Reproduce a small part of a research example and compare the observed learning behaviour.

A useful next step

Choose where to start.

Talk through your goals in a free 30-minute discovery call with Ted. Bring a task you want to improve and the tools you already use.

Book a discovery callExplore DoJo membership

Related learning

Explore the topic further.

Browse the course sample