Google Inc.

X Development LLC

Julian Ibarz

Sergey Levine

Alexander Irpan

Ian Goodfellow

Konstantinos Bousmalis

Yaroslav Bulatov

Vincent Olivier Vanhoucke

Kurt Konolige

Mrinal Kalakrishnan

Yunfei Bai

Sudheendra Vijayanarasimhan

Christopher Kenneth Harris

Kanury Kanishka Rao

Seyed Mohammad Khansari Zadeh

Peter Pastor Sampedro

Eric Jang

Paul Wohlhart

Sacha Christophe Arnoud

Kong Man Cheung

Marco Zennaro

Liron Yatziv

Alexander Herzog

Matthew Laurance Kelcey

Dmitry Kalashnikov

Kanishka Rao

Chris Harris

Sergey Vladimir Levine

Implementations disclosed herein relate to mitigating the reality gap through training a simulation-to-real machine learning model ('Sim2Real' model) using a vision-based robot task machine learning model. The vision-based robot task machine learning model can be, for example, a reinforcement learning (“RL”) neural network model (RL-network), such as an RL-network that represents a Q-function.

Mitigating reality gap through training a simulation-to-real model using a vision-based robot task model

Deep machine learning methods and apparatus related to semantic robotic grasping are provided. Some implementations relate to training a training a grasp neural network, a semantic neural network, and a joint neural network of a semantic grasping model. In some of those implementations, the joint network is a deep neural network and can be trained based on both: grasp losses generated based on grasp predictions generated over a grasp neural network, and semantic losses generated based on semantic predictions generated over the semantic neural network. Some implementations are directed to utilization of the trained semantic grasping model to servo, or control, a grasping end effector of a robot to achieve a successful grasp of an object having desired semantic feature(s).

Machine learning methods and apparatus for semantic robotic grasping

Asynchronous robotic control utilizing a trained critic network. During performance of a robotic task based on a sequence of robotic actions determined utilizing the critic network, a corresponding next robotic action of the sequence is determined while a corresponding previous robotic action of the sequence is still being implemented. Optionally, the next robotic action can be fully determined and/or can begin to be implemented before implementation of the previous robotic action is completed. In determining the next robotic action, most recently selected robotic action data is processed using the critic network, where such data conveys information about the previous robotic action that is still being implemented. Some implementations additionally or alternatively relate to determining when to implement a robotic action that is determined in an asynchronous manner.

Asynchronous robotic control using most recently selected robotic action data

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for off-policy evaluation of a control policy. One of the methods includes obtaining policy data specifying a control policy for controlling a source agent interacting with a source environment to perform a particular task; obtaining a validation data set generated from interactions of a target agent in a target environment; determining a performance estimate that represents an estimate of a performance of the control policy in controlling the target agent to perform the particular task in the target environment; and determining, based on the performance estimate, whether to deploy the control policy for controlling the target agent to perform the particular task in the target environment.

Off-policy control policy evaluation

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training an action selection neural network that is used to control a robotic agent interacting with a real-world environment.

Using simulation and domain adaptation for robotic control

Systems and methods for sequence transcription with neural networks are provided. More particularly, a neural network can be implemented to map a plurality of training images received by the neural network into a probabilistic model of sequences comprising P(S|X) by maximizing log P(S|X) on the plurality of training images. X represents an input image and S represents an output sequence of characters for the input image. The trained neural network can process a received image containing characters associated with building numbers. The trained neural network can generate a predicted sequence of characters by processing the received image.

Sequence transcription with deep neural networks

Systems and methods for updating geographic data based on a transaction are provided. In some aspects, one or more transaction records associated with a business are accessed from a memory. Each transaction record identifies a transaction time, geographic location data, and transaction information. A geocoded record of the business is selected to update, based on the geographic location data of the one or more transaction records. The selected geocoded record is updated based on at least one of the transaction time or the transaction information identified in the transaction records.

Updating geographic data based on a transaction

Growing community of inventors

Sunnyvale, CA, United States of America

Julian Ibarz