Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for data-driven robotic control. One of the methods includes maintaining robot experience data; obtaining annotation data; training, on the annotation data, a reward model; generating task-specific training data for the particular task, comprising, for each experience in a second subset of the experiences in the robot experience data: processing the observation in the experience using the trained reward model to generate a reward prediction, and associating the reward prediction with the experience; and training a policy neural network on the task-specific training data for the particular task, wherein the policy neural network is configured to receive a network input comprising an observation and to generate a policy output that defines a control policy for a robot performing the particular task.

Abstract: A computer-implemented method for preserving media streams may include (i) identifying a media stream transmitted by a client device to a server that hosts the media stream for access by additional devices, (ii) detecting that the server is expected to go offline, (iii) sending, in response to detecting that the server is expected to go offline, a message to the client device indicating that the server is expected to go offline, (iv) receiving, at an additional server, a request from the client device to host the media stream, and (v) in response to receiving the request, hosting the media stream at the additional server while ceasing to host the media stream at the server that is expected to go offline. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network used to select actions performed by an agent interacting with an environment by performing actions that cause the environment to transition states. One of the methods includes maintaining a replay memory storing a plurality of transitions; selecting a plurality of transitions from the replay memory; and training the neural network on the plurality of transitions, comprising, for each transition: generating an initial Q value for the transition; determining a scaled Q value for the transition; determining a scaled temporal difference learning target for the transition; determining an error between the scaled temporal difference learning target and the scaled Q value; determining an update to the current values of the Q network parameters; and determining an update to the current value of the scaling term.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for training machine learning models. One method includes obtaining a machine learning model, wherein the machine learning model comprises one or more model parameters, and the machine learning model is trained using gradient descent techniques to optimize an objective function; determining an update rule for the model parameters using a recurrent neural network (RNN); and applying a determined update rule for a final time step in a sequence of multiple time steps to the model parameters.

Abstract: A method of generating an output image having an output resolution of N pixels×N pixels, each pixel in the output image having a respective color value for each of a plurality of color channels, the method comprising: obtaining a low-resolution version of the output image; and upscaling the low-resolution version of the output image to generate the output image having the output resolution by repeatedly performing the following operations: obtaining a current version of the output image having a current K×K resolution; and processing the current version of the output image using a set of convolutional neural networks that are specific to the current resolution to generate an updated version of the output image having a 2K×2K resolution.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for data-driven robotic control. One of the methods includes maintaining robot experience data; obtaining annotation data; training, on the annotation data, a reward model; generating task-specific training data for the particular task, comprising, for each experience in a second subset of the experiences in the robot experience data: processing the observation in the experience using the trained reward model to generate a reward prediction, and associating the reward prediction with the experience; and training a policy neural network on the task-specific training data for the particular task, wherein the policy neural network is configured to receive a network input comprising an observation and to generate a policy output that defines a control policy for a robot performing the particular task.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an action selection policy neural network, wherein the action selection policy neural network is configured to process an observation characterizing a state of an environment to generate an action selection policy output, wherein the action selection policy output is used to select an action to be performed by an agent interacting with an environment. In one aspect, a method comprises: obtaining an observation characterizing a state of the environment subsequent to the agent performing a selected action; generating a latent representation of the observation; processing the latent representation of the observation using a discriminator neural network to generate an imitation score; determining a reward from the imitation score; and adjusting the current values of the action selection policy neural network parameters based on the reward using a reinforcement learning training technique.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for training machine learning models. One method includes obtaining a machine learning model, wherein the machine learning model comprises one or more model parameters, and the machine learning model is trained using gradient descent techniques to optimize an objective function; determining an update rule for the model parameters using a recurrent neural network (RNN); and applying a determined update rule for a final time step in a sequence of multiple time steps to the model parameters.

Abstract: A semiconductor photomultiplier module (20; 30; 40) comprises a first semiconductor chip (21; 31; 41) disposed in a first plane and comprising an array (21.1) of single-photon avalanche diodes (SPAD), a second semiconductor chip (22; 32; 42) disposed in a second plane and comprising a first part of an electronic read-out circuit, and a third semiconductor chip (26; 36; 46) comprising a second part of the electronic read-out circuit, wherein the first semiconductor chip (21; 31; 41) and the second semiconductor chip (22; 32; 42) are arranged in a stacked relationship and vertical electrical interconnects (23) are arranged to electrically interconnect the first semiconductor chip (21; 31; 41) with the second semiconductor chip (22; 32; 42).

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting actions to be performed by an agent to interact with an environment using an action selection neural network. In one aspect, a method comprises, at each time step in a sequence of time steps: generating a current representation of a state of a task being performed by the agent in the environment as of the current time step as a sequence of data elements; autoregressively generating a sequence of data elements representing a current action to be performed by the agent at the current time step; and after autoregressively generating the sequence of data elements representing the current action, causing the agent to perform the current action at the current time step.

Abstract: A computer-implemented method for preserving media streams may include (i) identifying a media stream transmitted by a client device to a server that hosts the media stream for access by additional devices, (ii) detecting that the server is expected to go offline, (iii) sending, in response to detecting that the server is expected to go offline, a message to the client device indicating that the server is expected to go offline, (iv) receiving, at an additional server, a request from the client device to host the media stream, and (v) in response to receiving the request, hosting the media stream at the additional server while ceasing to host the media stream at the server that is expected to go offline. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A method of generating an output image having an output resolution of N pixels×N pixels, each pixel in the output image having a respective color value for each of a plurality of color channels, the method comprising: obtaining a low-resolution version of the output image; and upscaling the low-resolution version of the output image to generate the output image having the output resolution by repeatedly performing the following operations: obtaining a current version of the output image having a current K×K resolution; and processing the current version of the output image using a set of convolutional neural networks that are specific to the current resolution to generate an updated version of the output image having a 2K×2K resolution.

Abstract: The present invention relates to a symmetrical and reversible-clutch mortise lock comprising a central follower, a pair of symmetrical lateral followers, each one located on one side of the central follower to be activated by square bars of handles, a spacer which makes the movement of the two lateral followers independent and a motor, which connects to an actuation lever and which is activated when an access code is validated. The lock further comprises a clutch housed and guided in an accumulator arm able to move linearly to be housed in both lateral followers such that the corresponding lateral follower, when activated by the motion of the door handle, pushes the clutch until it makes contact with the central follower and activates the latch of the lock.

Abstract: A method is proposed of training a neural network to generate action data for controlling an agent to perform a task in an environment. The method includes obtaining, for each of a plurality of performances of the task, one or more first tuple datasets, each first tuple dataset comprising state data characterizing a state of the environment at a corresponding time during the performance of the task; and a concurrent process of training the neural network and a discriminator network. The training process comprises a plurality of neural network update steps and a plurality of discriminator network update steps.

Abstract: A method of generating an output image having an output resolution of N pixels×N pixels, each pixel in the output image having a respective color value for each of a plurality of color channels, the method comprising: obtaining a low-resolution version of the output image; and upscaling the low-resolution version of the output image to generate the output image having the output resolution by repeatedly performing the following operations: obtaining a current version of the output image having a current K×K resolution; and processing the current version of the output image using a set of convolutional neural networks that are specific to the current resolution to generate an updated version of the output image having a 2K×2K resolution.

Abstract: Silicon-based photomultipliers (SiPMs) for reducing optical crosstalk effects in the SiPMs are provided. The SiPMs include macrocells. Each macrocell includes microcells, coupled in parallel, and a reading circuit coupled to an output of each macrocell. The microcells are arranged in the SiPM so that adjacent microcells belong to different macrocells. When a microcell performs a detection, the reading circuit of each macrocell having one or more microcells adjacent to the microcell that performed the detection is configured to disable its output signal during a predefined period of time. PET devices or systems and methods for reducing crosstalk effects are also provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for data-driven robotic control. One of the methods includes maintaining robot experience data; obtaining annotation data; training, on the annotation data, a reward model; generating task-specific training data for the particular task, comprising, for each experience in a second subset of the experiences in the robot experience data: processing the observation in the experience using the trained reward model to generate a reward prediction, and associating the reward prediction with the experience; and training a policy neural network on the task-specific training data for the particular task, wherein the policy neural network is configured to receive a network input comprising an observation and to generate a policy output that defines a control policy for a robot performing the particular task.

Abstract: A method of generating an output image having an output resolution of N pixels×N pixels, each pixel in the output image having a respective color value for each of a plurality of color channels, the method comprising: obtaining a low-resolution version of the output image; and upscaling the low-resolution version of the output image to generate the output image having the output resolution by repeatedly performing the following operations: obtaining a current version of the output image having a current K×K resolution; and processing the current version of the output image using a set of convolutional neural networks that are specific to the current resolution to generate an updated version of the output image having a 2K×2K resolution.

Abstract: The present invention relates to a symmetrical and reversible-clutch mortise lock comprising a central follower, a pair of symmetrical lateral followers, each one located on one side of the central follower to be activated by square bars of handles, a spacer which makes the movement of the two lateral followers independent and a motor, which connects to an actuation lever and which is activated when an access code is validated. The lock further comprises a clutch housed and guided in an accumulator arm able to move linearly to be housed in both lateral followers such that the corresponding lateral follower, when activated by the motion of the door handle, pushes the clutch until it makes contact with the central follower and activates the latch of the lock.

Abstract: A reinforcement learning system is proposed comprising a plurality of property detector neural networks. Each property detector neural network is arranged to receive data representing an object within an environment, and to generate property data associated with a property of the object. A processor is arranged to receive an instruction indicating a task associated with an object having an associated property, and process the output of the plurality of property detector neural networks based upon the instruction to generate a relevance data item. The relevance data item indicates objects within the environment associated with the task. The processor also generates a plurality of weights based upon the relevance data item, and, based on the weights, generates modified data representing the plurality of objects within the environment. A neural network is arranged to receive the modified data and to output an action associated with the task.