Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network system used to control an agent interacting with an environment to perform a specified task. One of the methods includes causing the agent to perform a task episode in which the agent attempts to perform the specified task; for each of one or more particular time steps in the sequence: generating a modified reward for the particular time step from (i) the actual reward at the time step and (ii) value predictions at one or more time steps that are more than a threshold number of time steps after the particular time step in the sequence; and training, through reinforcement learning, the neural network system using at least the modified rewards for the particular time steps.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network used to select actions to be performed by an agent interacting with an environment. One of the methods includes: obtaining data identifying a set of trajectories, each trajectory comprising a set of observations characterizing a set of states of the environment and corresponding actions performed by another agent in response to the states; obtaining data identifying an encoder that maps the observations onto embeddings for use in determining a set of imitation trajectories; determining, for each trajectory, a corresponding embedding by applying the encoder to the trajectory; determining a set of imitation trajectories by applying a policy defined by the neural network to the embedding for each trajectory; and adjusting parameters of the neural network based on the set of trajectories, the set of imitation trajectories and the embeddings.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating sequences of predicted observations, for example images. In one aspect, a system comprises a controller recurrent neural network, and a decoder neural network to process a set of latent variables to generate an observation. An external memory and a memory interface subsystem is configured to, for each of a plurality of time steps, receive an updated hidden state from the controller, generate a memory context vector by reading data from the external memory using the updated hidden state, determine a set of latent variables from the memory context vector, generate a predicted observation by providing the set of latent variables to the decoder neural network, write data to the external memory using the latent variables, the updated hidden state, or both, and generate a controller input for a subsequent time step from the latent variables.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for augmenting neural networks with an external memory using reinforcement learning. One of the methods includes providing an output derived from the system output portion of the neural network output as a system output in the sequence of system outputs; selecting a memory access process from a predetermined set of memory access processes for accessing the external memory from the reinforcement learning portion of the neural network output; writing and reading data from locations in the external memory in accordance with the selected memory access process using the differentiable portion of the neural network output; and combining the data read from the external memory with a next system input in the sequence of system inputs to generate a next neural network input in the sequence of neural network inputs.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for augmenting neural networks with an external memory. One of the systems includes a sparse memory access subsystem that is configured to perform operations comprising generating a sparse set of reading weights that includes a respective reading weight for each of the plurality of locations in the external memory using the read key, reading data from the plurality of locations in the external memory in accordance with the sparse set of reading weights, generating a set of writing weights that includes a respective writing weight for each of the plurality of locations in the external memory, and writing the write vector to the plurality of locations in the external memory in accordance with the writing weights.

Abstract: Systems, methods, and apparatus, including computer programs encoded on a computer storage medium, related to associative long short-term memory (LSTM) neural network layers configured to maintain N copies of an internal state for the associative LSTM layer, N being an integer greater than one. In one aspect, a system includes a recurrent neural network including an associative LSTM layer, wherein the associative LSTM layer is configured to, for each time step, receive a layer input, update each of the N copies of the internal state using the layer input for the time step and a layer output generated by the associative LSTM layer for a preceding time step, and generate a layer output for the time step using the N updated copies of the internal state.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for augmenting neural networks with an external memory. One of the systems includes a memory interface subsystem that is configured to perform operations comprising determining a respective content-based weight for each of a plurality of locations in an external memory; determining a respective allocation weight for each of the plurality of locations in the external memory; determining a respective final writing weight for each of the plurality of locations in the external memory from the respective content-based weight for the location and the respective allocation weight for the location; and writing data defined by the write vector to the external memory in accordance with the final writing weights.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for augmenting neural networks to generate additional outputs. One of the systems includes a neural network and a sequence processing subsystem, wherein the sequence processing subsystem is configured to perform operations comprising, for each of the system inputs in a sequence of system inputs: receiving the system input; generating an initial neural network input from the system input; causing the neural network to process the initial neural network input to generate an initial neural network output for the system input; and determining, from a first portion of the initial neural network output for the system input, whether or not to cause the neural network to generate one or more additional neural network outputs for the system input.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for augmenting neural networks with an external memory. One of the methods includes providing an output derived from a first portion of a neural network output as a system output; determining one or more sets of writing weights for each of a plurality of locations in an external memory; writing data defined by a third portion of the neural network output to the external memory in accordance with the sets of writing weights; determining one or more sets of reading weights for each of the plurality of locations in the external memory from a fourth portion of the neural network output; reading data from the external memory in accordance with the sets of reading weights; and combining the data read from the external memory with a next system input to generate the next neural network input.

Abstract: Providing quality management by validating a bill of materials in event-based product manufacturing. Some of the disclosed embodiments include a system and method for validating materials during manufacture of a product. Also disclosed are a method of validating that acceptable materials are being used on a machine during manufacture of a product, and one or more computer-readable media for storing a data structure representing a product specification in an event-based manufacturing system.

Abstract: Providing quality management and intelligent manufacturing with labels and smart tags in event-based product manufacturing. Some of the disclosed embodiments include a system, method, and computer-readable media for storing, during a process, data associated with a material. Also disclosed are a method of collecting, storing, and reporting machine productivity, waste, and delay information on an event basis in a manufacturing system, a method of capturing and storing material history, a method of automating tracking of positions of components used in a process and correlating portions of a component with production problems, an improved inventory management system, and a method of tracking and recording actions of specific operators of a process performed by a machine.

Abstract: Communication between machines in an event-based manufacturing system. In an event-based manufacturing system, material-specific data obtained for a material produced during a first manufacturing operation is used to govern a second manufacturing operation to decrease the likelihood of at least a delay event, a waste event, or poor quality of the product.

Abstract: Integrating event-based production information with financial and purchasing systems in product manufacturing. Some of the disclosed embodiments include a method of tracking production information, a method of analyzing event-based production information to provide financial reports, a method of adapting an online production documentation system to provide financial report data, and a production documentation system. Also disclosed are computer-readable media for storing a data structure representing an event-based delay or waste record populated during manufacturing, and a method of automatically generating an alert and providing an associated financial report in a process for manufacturing a product associated with an event-based information system.

Abstract: A high frequency flexure-based dynamometer (10) for measuring vibrations to use in determining cutting forces in a tool. The dynamometer device (10) may operate within a preselected high frequency range while measuring cutting forces less than about 1 N. The dynamometer (10) may include two coupled flexures that interact to produce vibration modes at the edge of a selected bandwidth of interest. These modes may produce a frequency response function within the desired frequency band that has a magnified response and is substantially constant. The dynamometer (10) may include a workpiece (38) mounted to one of the two flexures (12) and a one or more precision accelerometers (30) mounted to the first or second flexures. Finite element analysis may be used to optimize the flexure design.

Abstract: The reproduction of a requested source document in a requested available form (including electronic, print, audio and Braille) is disclosed. At a server, for each one of a plurality of documents at least one access pathway is applied to a marked-up form of the document. The access pathways define discrete parts of the document. A fragment of the marked-up document is generated for each said access pathway for each available form. A requested one or more parts of a source document is generated in a requested form from the respective stored fragments. The fragments are transmitted to the requesting customer for local reproduction.

Abstract: The present invention relates to an apparatus for packaging contact lenses. The apparatus includes a compliant head, improved heater and improved heat seal die.

Abstract: A method for predicting a tool point response of a spindle-holder-tool assembly to be used for high-speed machining applications is provided. The method includes determining direct and cross displacement-to-force receptances of a standard holder secured in the spindle based upon measurements of displacement-to-force taken at different rotating speeds of the spindle. The method also includes determining direct receptances at a free end of the standard holder based upon the determined direct and cross displacement-to-force receptances of the standard holder. Additionally, the method includes performing an inverse receptance coupling to simulate a decomposition of the standard holder into multiple subassemblies, the subassemblies including a spindle-holder base subassembly and an extended holder subassembly. The method further includes determining spindle-holder base subassembly receptances based upon the simulated decomposition of the standard holder.

Abstract: A preferred process for degassing a mold piece includes inserting the mold piece into an internal volume within a canister, sealing the internal volume, drawing a vacuum in the internal volume, and introducing into the internal volume a gaseous fluid suitable for degassing the mold piece.

Abstract: A system (5) for producing a document is disclosed. A repository (16) stores documents in a marked-up form according to one or more mark-up schemas. The schemas are adapted to make explicit structural information contained a document. A document format store (22) stores formats that are user selectable. A document production processor (29, 31, 32) generates a user-requested formatted document from the respective marked-up form using a user-selected said format that retains said implicit structural information.

Abstract: A water supply system for supplying water to a building including a water tank (13) for collecting water from the roof (10) and gutters (12) of the building (11); a water level sensor means (21) for sensing the level of water in the water tank; a tank water system (22) for supplying water from the water tank (13); a water pump (23) for pumping water from the water tank (13) along the tank water system (22); a non-return valve (25) downstream of the pump in the tank water system; a pressure sensor means (26) for sensing water pressure of water within the tank water system, said pressure sensor means (26) is located downstream of the non-return valve (25); a mains water system for supplying mains water; a three port valve (35) having a first port (34) connected to the tank water system downstream from the pressure sensor means (26), a second port (37) connected to the mains water system and a third port (41) connected to the building's water reticulation system; wherein when the first (34) and third port (41) a