نتائج البحث
Hard Questions: Is Spending Time on Social Media Bad for Us? - meta.com
Hard Questions: Is Spending Time on Social Media Bad for Us? meta.com
Block-sparse GPU kernels
We’re releasing highly-optimized GPU kernels for an underexplored class of neural network architectures: networks with block-sparse weights. Depending on the chosen sparsity, these kernels can run orders of magnitude faster than cuBLAS or cuSPARSE. We’ve used them to attain state-of-the-art results in text sentiment analysis and generative modeling of text and images.
Block-sparse GPU kernels
We’re releasing highly-optimized GPU kernels for an underexplored class of neural network architectures: networks with block-sparse weights. Depending on the chosen sparsity, these kernels can run orders of magnitude faster than cuBLAS or cuSPARSE. We’ve used them to attain state-of-the-art results in text sentiment analysis and generative modeling of text and images.
Facebook’s 2017 Year In Review - meta.com
Facebook’s 2017 Year In Review meta.com
Introducing Messenger Kids, a New App For Families to Connect - meta.com
Introducing Messenger Kids, a New App For Families to Connect meta.com
Hard Questions: So Your Kids Are Online, But Will They Be Alright? - meta.com
Hard Questions: So Your Kids Are Online, But Will They Be Alright? meta.com
Facebook Social Good Forum: Announcing New Tools and Initiatives for Communities to Help Each Other - meta.com
Facebook Social Good Forum: Announcing New Tools and Initiatives for Communities to Help Each Other meta.com
A look back at 10 years of the Amazon Kindle - About Amazon
A look back at 10 years of the Amazon Kindle About Amazon
Introducing our first airplane: Amazon One - About Amazon
Introducing our first airplane: Amazon One About Amazon
Send Money to Friends in Messenger – Now in Euros and British Pounds - meta.com
Send Money to Friends in Messenger – Now in Euros and British Pounds meta.com
Amazon’s Joining Forces pledge - About Amazon
Amazon’s Joining Forces pledge About Amazon
Semi – Electric Semi Truck - Tesla
Semi – Electric Semi Truck Tesla
Update on Our Advertising Transparency and Authenticity Efforts - meta.com
Update on Our Advertising Transparency and Authenticity Efforts meta.com
Learning a hierarchy
We’ve developed a hierarchical reinforcement learning algorithm that learns high-level actions useful for solving a range of tasks, allowing fast solving of tasks requiring thousands of timesteps. Our algorithm, when applied to a set of navigation problems, discovers a set of high-level actions for walking and crawling in different directions, which enables the agent to master new navigation tasks quickly.
Learning a hierarchy
We’ve developed a hierarchical reinforcement learning algorithm that learns high-level actions useful for solving a range of tasks, allowing fast solving of tasks requiring thousands of timesteps. Our algorithm, when applied to a set of navigation problems, discovers a set of high-level actions for walking and crawling in different directions, which enables the agent to master new navigation tasks quickly.
glamazon at Amazon: eighteen years of change - About Amazon
glamazon at Amazon: eighteen years of change About Amazon
Generalizing from simulation
Our latest robotics techniques allow robot controllers, trained entirely in simulation and deployed on physical robots, to react to unplanned changes in the environment as they solve simple tasks. That is, we’ve used these techniques to build closed-loop systems rather than open-loop ones as before.
Generalizing from simulation
Our latest robotics techniques allow robot controllers, trained entirely in simulation and deployed on physical robots, to react to unplanned changes in the environment as they solve simple tasks. That is, we’ve used these techniques to build closed-loop systems rather than open-loop ones as before.
Meta-learning for wrestling
We show that for the task of simulated robot wrestling, a meta-learning agent can learn to quickly defeat a stronger non-meta-learning agent, and also show that the meta-learning agent can adapt to physical malfunction.
Meta-learning for wrestling
We show that for the task of simulated robot wrestling, a meta-learning agent can learn to quickly defeat a stronger non-meta-learning agent, and also show that the meta-learning agent can adapt to physical malfunction.