Neural networks are part of many contemporary NLP systems, yet their
empirical successes come at the price of vulnerability to adversarial attacks.
Previous work has used adversarial training and data augmentation to partially
Our waking and sleeping lives are punctuated by fragments of recalled memories: a sudden connection in the shower between seemingly disparate thoughts, or an ill-fated choice decades ago that haunts us as we struggle to fall asleep.
DeepMindIn our new blog post, we review how brains replay experiences to strengthen memories, and how researchers use the same principle to train better AI systems
This paper introduces R2D3, an agent that makes efficient use of
demonstrations to solve hard exploration problems in partially observable
environments with highly variable initial conditions. We also introduce a suite
of eight tasks that combine...
Multi-object image datasets with ground-truth segmentation masks and generative factors. - deepmind/multi_object_datasets
DeepMindWe’ve released datasets for scene decomposition & representation learning research: github.com/deepmind/multi….
Each image comes with ground-truth masks & features for all objects. We hope to facilitate unsupervised learning in this area, building on models like MONet & IODINE. pic.twitter.com/gLAX902KWb
OpenSpiel is a collection of environments and algorithms for research in
general reinforcement learning and search/planning in games. OpenSpiel supports
n-player (single- and multi- agent) zero-sum, cooperative and general-sum,
We apply three machine learning strategies to optimize the atomic cooling
processes utilized in the production of a Bose-Einstein condensate (BEC). For
the first time, we optimize both laser cooling and evaporative cooling
DeepMindIn collaboration w/ @ox_ultracold & @FelixHill84, we developed machine learning techniques to optimise the production of a Bose-Einstein condensate, a quantum-mechanical state of matter that can be used to test predictions of theories of many-body physics
We research and build safe AI systems that learn how to solve problems and advance scientific discovery for all. Explore our work: deepmind.com/research
DeepMindApplications are now open for our 2020 Research Scientist and Engineering internships!
We offer hands-on experience working collaboratively on projects that push the frontiers of AI and science across a wide variety of teams.
Find out more or apply
Read the latest articles and stories from DeepMind and find out more about our latest breakthroughs in cutting-edge AI research.
DeepMindWe’re excited to release episodes 1 - 4 of the #DMpodcast! Get the inside track on some of the big questions and challenges the field is wrestling with today. No need to be an expert - the amazing @FryRsquared speaks to the people behind the science.
Spriteworld: a flexible, configurable python-based reinforcement learning environment - deepmind/spriteworld
DeepMindWe're open sourcing Spriteworld, a flexible, procedural reinforcement learning environment: github.com/deepmind/sprit…
Spriteworld is a 2-dimensional arena with movable objects, and is particularly well-suited for small-scale experiments with limited computational resources. pic.twitter.com/Nw6Zfgg4ht
Can an arbitrarily intelligent reinforcement learning agent be kept under
control by a human user? Or do agents with sufficient intelligence inevitably
find ways to shortcut their reward signal? This question impacts how far
This paper introduces the Behaviour Suite for Reinforcement Learning, or
bsuite for short. bsuite is a collection of carefully-designed experiments that
investigate core capabilities of reinforcement learning (RL) agents with two
DeepMindWe built bsuite to do two things:
1. Offer clear, informative, and scalable experiments that capture key issues in RL
2. Study agent behaviour through performance on shared benchmarks
You can get started with bsuite in this colab