Rank #2121h ago
ICLR 2026 awards Test of Time to DCGAN and continuous control papers
LIZG
4 top authorsICLR 2026 awards its Test of Time Award to two 2016 papers: 'Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks' (DCGAN) by Alec Radford, Luke Metz, and Soumith Chintala, advancing unsupervised representation learning; and 'Continuous Control with Deep Reinforcement Learning' by Timothy P. Lillicrap and colleagues, impacting robotics control methods. Google Research announced the honors, with a dedicated DCGAN talk at the conference.
First post
Why it matters
Luke Metz (AI researcher) coauthored the DCGAN paper at indico Research before his later roles at Google Brain and OpenAI.
The continuous control paper demonstrated deep reinforcement learning across more than 20 simulated robotics tasks including legged locomotion.
3 more posts
GO
