Manipulation, locomotion, sim-to-real transfer and autonomous driving: learning systems that have to survive physics.
An encoder (optical system) maps objects to noiseless images, which noise corrupts into measurements. Our information estimator uses only these noisy measurements and a noise model to quantify how well measurements distinguish objects.
In this post, I’ll introduce a reinforcement learning (RL) algorithm based on an “alternative” paradigm: divide and conquer . Unlike traditional methods, this algorithm is not based on temporal difference (TD) learning (which has scalability challenges ), and scales well to long-horizon tasks.
We’re powering an era of physical agents — enabling robots to perceive, plan, think, use tools and act to better solve complex, multi-step tasks.
We’re introducing an efficient, on-device robotics model with general-purpose dexterity and fast task adaptation.
We’ve trained a pair of neural networks to solve the Rubik’s Cube with a human-like robot hand. The neural networks are trained entirely in simulation, using the same reinforcement learning code as OpenAI Five paired with a new technique called Automatic Domain Randomization (ADR).
We hosted the first OpenAI Robotics Symposium on April 27, 2019.
We’ve developed an energy-based model that can quickly learn to identify and generate instances of concepts, such as near, above, between, closest, and furthest, expressed as sets of 2d points. Our model learns these concepts after only five demonstrations.
We’ve trained a human-like robot hand to manipulate physical objects with unprecedented dexterity.
