recore
ReCoRe: Regularized Contrastive Representation Learning of World Model
Poudel, Rudra P. K., Pandya, Harit, Liwicki, Stephan, Cipolla, Roberto
While recent model-free Reinforcement Learning (RL) methods have demonstrated human-level effectiveness in gaming environments, their success in everyday tasks like visual navigation has been limited, particularly under significant appearance variations. This limitation arises from (i) poor sample efficiency and (ii) over-fitting to training scenarios. To address these challenges, we present a world model that learns invariant features using (i) contrastive unsupervised learning and (ii) an intervention-invariant regularizer. Learning an explicit representation of the world dynamics i.e. a world model, improves sample efficiency while contrastive learning implicitly enforces learning of invariant features, which improves generalization. However, the naive integration of contrastive loss to world models fails due to a lack of supervisory signals to the visual encoder, as world-model-based RL methods independently optimize representation learning and agent policy. To overcome this issue, we propose an intervention-invariant regularizer in the form of an auxiliary task such as depth prediction, image denoising, etc., that explicitly enforces invariance to style-interventions. Our method outperforms current state-of-the-art model-based and model-free RL methods and significantly on out-of-distribution point navigation task evaluated on the iGibson benchmark. We further demonstrate that our approach, with only visual observations, outperforms recent language-guided foundation models for point navigation, which is essential for deployment on robots with limited computation capabilities. Finally, we demonstrate that our proposed model excels at the sim-to-real transfer of its perception module on Gibson benchmark.
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.14)
- Europe > Austria > Vienna (0.14)
- Information Technology > Artificial Intelligence > Representation & Reasoning (1.00)
- Information Technology > Artificial Intelligence > Cognitive Science > Problem Solving (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.93)
- Information Technology > Artificial Intelligence > Machine Learning > Learning Graphical Models > Undirected Networks > Markov Models (0.46)
ReCore review: The first Xbox Play Anywhere game is a beautiful chore
The biggest disappointment reviewing ReCore is how strong it starts. You awake from stasis on the distant planet of Far Eden expecting to find a lush, terraformed planet capable of sustaining human life, 200 years in the making. Instead you find sand dunes. Miles upon miles of sand dunes, broken only by craggy brown rocks and the occasional half-built skyscraper. It falls to you to figure out what happened, traipsing off across the desert in search of answers with your trusty robo-dog companion Mack.
Microsoft Xbox E3 2016 Live Stream: Watch The Xbox One S, 'ReCore' And 'Gears Of War 4' Reveals Here
Microsoft kicks off a busy Monday on the first full day of E3 2016. With the Xbox One trailing Sony's PS4 in terms of sales, Microsoft needs to get players excited about the console. There could be a slimmer version of the Xbox One, along with exclusive games. The Microsoft E3 2016 live stream begins Monday at 9:30 a.m. The big rumor ahead of E3 2016 surrounded a slim version of the current Xbox One with double the memory.
Microsoft E3 2016 Live Stream: Watch The Xbox One S, 'ReCore' And 'Gears Of War 4' Reveals Here
Microsoft kicks off a busy Monday on the first full day of E3 2016. With the Xbox One trailing Sony's PS4 in terms of sales, Microsoft needs to get players excited about the console. There could be a slimmer version of the Xbox One, along with exclusive games. The Microsoft E3 2016 live stream begins Monday at 9 a.m. The big rumor ahead of E3 2016 surrounded a slim version of the current Xbox One with double the memory.