Synthesizing Programs for Images using Reinforced Adversarial Learning
This session was not filmed.
The code for Reinforcement Learning agents trained in this paper was open sourced last month here:
https://github.com/deepmind/spiral
This paper proposes a very interesting application of RL on generating images. Current methods that combine deep learning and renderers are limited by hand-crafted likelihood or distance functions, which creates a need for large amounts of supervision, or difficulties in scaling their inference algorithms to richer datasets.
To mitigate these issues, SPIRAL is an adversarially trained agent that generates a program which is executed by a graphics engine to interpret and sample images. You can see it performing here: https://www.youtube.com/watch?v=iSyvwAwa7vk&feature=youtu.be
YOU MAY ALSO LIKE:
- Reinforcement Learning & Creative Applications (SkillsCast recorded in August 2019)
- Brian Sletten's Data Science with Python Workshop (in London on 18th - 20th November 2019)
- Fast Track to Machine Learning with Louis Dorard (in London on 2nd - 4th December 2019)
- Scala eXchange London 2019 (in London on 12th - 13th December 2019)
- Practical ML 2020 (in London on 2nd - 3rd July 2020)
- The Sonic Contender (in London on 28th October 2019)
- Solandra Hands-On Tutorial & Emergent Behaviour In Insects (in London on 28th October 2019)
- The Five Stages of Data: A Holistic Approach to Data Analytics and BI (SkillsCast recorded in October 2019)
- Implementing Clean Architecture in Flutter using BLoC (SkillsCast recorded in October 2019)
Thanks to our sponsors
Synthesizing Programs for Images using Reinforced Adversarial Learning
Ali Chaudhry
Ali is a self-taught programmer, currently doing a PhD in Artificial Intelligence and Education at University College London. His research focuses on applying human-centered learning science theories on Deep Reinforcement Learning algorithms. These days he's obsessed with GANs and Variational Autoencoders.