Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network having a plurality of network parameters and being configured to process an input data item to generate a feature representation comprising a values for each of a plurality of features of the input data item.

Abstract: The invention provides for systems, methods, and compositions for targeting nucleic acids. In particular, the invention provides non-naturally occurring or engineered DNA-targeting systems comprising a novel DNA-targeting CRISPR effector protein and at least one targeting nucleic acid component like a guide RNA. Aspects of the invention in particular relate to Cpf1 mutants having altered PAM specificity.

Abstract: Various embodiments set forth systems and techniques for determining expandability of a budding design. The techniques include receiving a building geometry indicating a size of a building and a shape of the building; receiving building placement data indicating a placement of the building within a site; identifying one or more remaining spaces within the site based at least on the size of the building, the shape of the building, and the placement of the building within the site; for each remaining space of the one or more remaining spaces: determining whether the remaining space is a usable space; based on the determining whether the remaining space is a usable space, generating a space expansion value corresponding to the remaining space; and computing, based on space expansion values corresponding to the one or more remaining spaces, one or more aggregate expansion values associated with the building geometry and the building placement.

Abstract: A design engine is configured to automatically generate designs for multi-family housing projects that simultaneously meet local construction regulations while also meeting specific financial targets. A design generator within the design engine generates a first generation of design options that reflect historical design trends. A design evaluator within the design engine then generates design metrics that quantify various attributes of the different design options. The design generator identifies a subset of the design options that optimally balance some or all of the various design metrics, and then generates a subsequent generation of design options that includes design features derived from the subset of design options.

Abstract: A design engine is configured to automatically generate designs for multi-family housing projects that simultaneously meet local construction regulations while also meeting specific financial targets. The design engine includes a design analyzer that generates design trends based on a database of historical designs. A feature extractor within the design analyzer performs a geometrical analysis of the historical designs to generate a library of features. A feature parameterizer within the design analyzer performs a parameterization process to generate one or more parametric values for each feature. A parameter correlator within the design analyzer compares different subsets of parametric values to identify relationships between those values. A trend aggregator identifies relationships that occur across multiple historical designs to determine a set of design trends. The design engine generates design options using the design trends.

Abstract: One embodiment sets forth a computer-implemented method for measuring productivity in buildings and workplaces. The method includes receiving workstation data for a plurality of workstations for the building and operation data for a plurality of operations for the building; generating a building layout specifying, for each workstation of the plurality of workstations, a respective location of the workstation; generating, for each operation of the plurality of operations, a respective productivity value based on the workstation data, the operation data, and the locations of the plurality of workstations; and calculating, based on the productivity values of the plurality of operations, one or more overall productivity values associated with the building layout.

Abstract: Various embodiments set forth systems and techniques for generating work condition values for a building layout. The techniques include receiving a building layout specifying, for each workspace of a plurality of workspaces included in the workplace, a respective location of the workspace; selecting one or more work condition elements from a plurality of work condition elements based at least on the plurality of workspaces; for each work condition element of the one or more work condition elements: evaluating the plurality of workspaces based on the work condition element; based on the evaluating the plurality of workspaces, generating an element value corresponding to the work condition element; and computing, based on element values corresponding to the one or more work condition elements, an overall work condition value associated with the building layout.

Abstract: Various embodiments set forth systems and techniques for assessing sustainability of a building design. The techniques include receiving a building design for a budding and sustainability certification data describing a plurality of sustainability certification systems; selecting one or more particular sustainability certification systems from the plurality of sustainability certification systems; computing a respective sustainability certification value for each sustainability certification system of the one or more particular sustainability certification systems; and performing one or more building design operations based on sustainability certification system values for the one or more particular sustainability certification systems.

Abstract: A design simulator executes a shape parameter engine that generates geometry corresponding to parking lots. The shape parameter engine implements a library of shapes, action markers, and construction rules to generate designs for parking lots. Each shape is a geometric the that can populate a design surface. Each action marker signifies an action that can occur at a specific location on the design surface. Each construction rule describes how shapes should be placed on the design surface relative to one or more action markers. The shape parameter generator tiles the design surface using the library of shapes, action markers, and construction rules to generate valid parking lot designs.

Abstract: A design engine automatically generates designs for multi-family housing projects that simultaneously meet local construction regulations while also meeting specific financial targets. The design engine includes a design analyzer, a site analyzer, a design generator, and a design evaluator. The design analyzer generates design trends based on a historical database of designs. The site analyzer generates design criteria based on relevant construction regulations. The design generator generates design options that reflect the design trends while also adhering to the construction regulations. The design evaluator then analyzes the design options and generates various design metrics. Based on the design metrics, the design generator generates additional design options that better meet the design criteria.

Abstract: A computer-implemented method for computationally determining ventilation efficiency when generating a building design comprises: generating a first three-dimensional (3D) mesh based on a first 3D building model; performing a first fluid dynamic computer simulation based on the first 3D mesh and first environmental data associated with the first 3D building model to generate a first output data set; and computing, based on the first output data set, a first value for a ventilation performance metric that is associated with the first 3D building model.

Abstract: A design simulator executes a rigid body simulation to position design elements within various design options according to configurations that meet design criteria while also promoting feasibility and desirability. The design simulator assigns physical properties to each design element, including a mass value and a velocity value. The design simulator then simulates rigid body interactions between design elements over a sequence of simulation steps. The design simulator stops the rigid body simulation when no collisions between design elements are detected or when other termination criteria are met. Once the rigid body simulation is complete, the final positions of each design element relative to one another represent a specific configuration of design elements for the corresponding design option.

Abstract: A design engine is configured to automatically generate and position design elements within designs for a multi-family housing project that corresponds to a parcel of land. Each design simultaneously meets local construction regulations while also meeting specific financial targets. The design engine generates a site model that reflects geometric attributes of the parcel of land, a building model that describes one or more buildings configured to occupy the parcel of land, and a parking lot model that describes a parking lot configured to occupy a remaining portion of the parcel of land. A design simulator executes various types of physical simulations with the site model, the building model, and/or the parking lot model to generate a vast number of different design options having widely varying configurations of design elements. The design engine evaluates these design options to identify those which balance compliance with construction regulations and achievement of financial targets.

Abstract: A computer-implemented method for generating a building design for a building site having a complex shape comprises: generating a first group of processing areas from a plurality of processing areas associated with the building design and a second group of processing areas from the plurality of processing areas; generating a first building footprint leg that is sized to include the first group of processing areas and a second building footprint leg that is sized to include the second group of processing areas; at a pivot location, joining the first building footprint leg and the second building footprint leg at an aperture angle; performing a numerical simulation to position the first group of processing areas in the first building footprint leg and the second group of processing areas in the second building footprint leg to generate a candidate building design for the building site; and rendering the candidate building design.

Abstract: One or more embodiments disclose techniques for generating a layout for a physical space or a building. The techniques include generating a layout based on a floorplan of a physical space, generating a plurality of spatial unit grids corresponding to a plurality of spatial units to be placed in the physical space, identifying a placement for a first spatial unit grid in the plurality of spatial unit grids within the layout by matching a corner cell in the first spatial unit grid with a given available cell in the layout, generating a score associated with the placement for the first spatial unit grid based on one or more placement parameters that define placement constraints for a first spatial unit included in the plurality of spatial units and corresponding to the first spatial unit grid, and placing the first spatial unit grid in the layout based on the score.

Abstract: The present disclosure provides for systems, methods, and compositions for targeting and editing nucleic acids. In particular, the present disclosure provides non-naturally occurring or engineered RNA-targeting systems comprising a RNA-targeting Cas13 protein, at least one guide molecule, and at least one adenosine deaminase protein or catalytic domain thereof.

Abstract: A workflow engine processes a work problem to generate solutions for the work problem comprising a plurality of related optimization problems. The work problem may comprise a second optimization problem that is dependent on a first optimization problem, such that at least one solution for the first optimization problem is to be utilized as an initial solution for the second optimization problem. The workflow engine generates and stores a branch object for each optimization problem, each branch object specifying a solver engine assigned for processing the optimization problem and dependency information indicating a dependency relationship between the optimization problem and another optimization problem. The workflow engine processes the work problem based on the branch objects by initiating each solver engine to perform optimization operations on the assigned optimization problem based on the corresponding branch object to generate one or more solutions for the assigned optimization problem.

Abstract: An endovascular treatment device for use in a vessel during a therapy procedure includes a catheter and a vessel occluder coupled to the catheter. The occluder includes a construction of elastic filaments that are fixed to the outer surface of the catheter. The filaments extend radially and distally outward from their proximal ends and then invert and are fixed back at the catheter proximal of the distal end of the catheter. The occluder can be elastically deformed to fit vessels of various sizes. The catheter includes a lumen through which a therapeutic agent is delivered beyond the occluder.

Abstract: Techniques for broadcasting notifications with low latency are provided. A set of recipients for a notification of a content item is determined. The notification is stored in a second notifications storage that is separate from a first notifications storage. A request for one or more notifications is received over a computer network from a client device that is associated with a particular recipient in the set of recipients. In response to receiving the request, first notification data is retrieved from the first notifications storage, second notification data that includes the notification is retrieved from the second notifications storage. The first notification data is combined with the second notification data to generate combined notification data. The combined notification data is transmitted over the computer network to the client device and is stored in the first notifications storage. The notification may be removed from the first notifications storage.

Abstract: A collapsible ballistic shield which provides a degree of protection against ballistic threats. The shield includes a head plate, an upper torso plate, a lower torso plate, an upper lock-and-release hinge joint, a lower lock-and-release hinge joint, a primary handle, and a strap-mounting fastener. The head plate is positioned opposite to the lower torso plate, across the upper torso plate to outline a shape of a body. The head plate is pivotably attached to the upper torso plate by the upper lock-and-release hinge joint. The lower torso plate is pivotably attached to the upper torso plate by the lower lock-and-release hinge joint. Resultantly, the shield may be collapsed, expanded, and broken down. The primary handle provides a grasping point and is mounted to a rear surface of the upper torso plate. The strap-mounting fastener allows the user to attach the shield to external structures.