نتائج البحث
How AI training scales
We’ve discovered that the gradient noise scale, a simple statistical metric, predicts the parallelizability of neural network training on a wide range of tasks. Since complex tasks tend to have noisier gradients, increasingly large batch sizes are likely to become useful in the future, removing one potential limit to further growth of AI systems. More broadly, these results show that neural network training need not be considered a mysterious art, but can be rigorized and systematized.
Adventure Awaits! No time to waste!
Assalamu Alaykum Husna Travellers! Welcome to the Husna Family! We’re so happy to have you! My name is Sobia Hussain and I’ll be your exclusive Husna Travel & Excursion Representative. You’ll be hearing from me weekly with all the exciting adventures and exploration awaiting you in the Bahamas,...
Quantifying generalization in reinforcement learning
We’re releasing CoinRun, a training environment which provides a metric for an agent’s ability to transfer its experience to novel situations and has already helped clarify a longstanding puzzle in reinforcement learning. CoinRun strikes a desirable balance in complexity: the environment is simpler than traditional platformer games like Sonic the Hedgehog but still poses a worthy generalization challenge for state of the art algorithms.
How Are We Doing at Enforcing Our Community Standards? - meta.com
How Are We Doing at Enforcing Our Community Standards? meta.com
Plan online, learn offline: Efficient learning and exploration via model-based control
Plan online, learn offline: Efficient learning and exploration via model-based control
Reinforcement learning with prediction-based rewards
We’ve developed Random Network Distillation (RND), a prediction-based method for encouraging reinforcement learning agents to explore their environments through curiosity, which for the first time exceeds average human performance on Montezuma’s Revenge.
Reinforcement learning with prediction-based rewards
We’ve developed Random Network Distillation (RND), a prediction-based method for encouraging reinforcement learning agents to explore their environments through curiosity, which for the first time exceeds average human performance on Montezuma’s Revenge.
New Technology to Fight Child Exploitation - meta.com
New Technology to Fight Child Exploitation meta.com
Learning complex goals with iterated amplification
We’re proposing an AI safety technique called iterated amplification that lets us specify complicated behaviors and goals that are beyond human scale, by demonstrating how to decompose a task into simpler sub-tasks, rather than by providing labeled data or a reward function. Although this idea is in its very early stages and we have only completed experiments on simple toy algorithmic domains, we’ve decided to present it in its preliminary state because we think it could prove to be a scalable a...
Learning complex goals with iterated amplification
We’re proposing an AI safety technique called iterated amplification that lets us specify complicated behaviors and goals that are beyond human scale, by demonstrating how to decompose a task into simpler sub-tasks, rather than by providing labeled data or a reward function. Although this idea is in its very early stages and we have only completed experiments on simple toy algorithmic domains, we’ve decided to present it in its preliminary state because we think it could prove to be a scalable a...
Fighting Election Interference in Real Time - meta.com
Fighting Election Interference in Real Time meta.com
FFJORD: Free-form continuous dynamics for scalable reversible generative models
FFJORD: Free-form continuous dynamics for scalable reversible generative models
Amazon Global Press Center - About Amazon
Amazon Global Press Center About Amazon
OpenAI Scholars 2018: Final projects
Our first cohort of OpenAI Scholars has now completed the program.
OpenAI Scholars 2018: Final projects
Our first cohort of OpenAI Scholars has now completed the program.
On Our Way to Lower Emissions and 100% Renewable Energy - meta.com
On Our Way to Lower Emissions and 100% Renewable Energy meta.com
Removing Myanmar Military Officials From Facebook - meta.com
Removing Myanmar Military Officials From Facebook meta.com
The International 2018: Results
OpenAI Five lost two games against top Dota 2 players at The International in Vancouver this week, maintaining a good chance of winning for the first 20–35 minutes of both games.