Abstract: A method and system control a water distribution network. A database is maintained of prior states based on a residential water demand, a tank level, and an energy tariff. A current state of the water distribution network is determined. Rewards are determined and include a tank level constraint, an energy cost, and a toggle count. A query based model is used to determine a set of control points used to control a first prior state. An RL agent is trained based on the prior states and rewards. The RL agent determines a control setpoint (that changes the pump speed) that maintains the tank level, minimizes the energy cost, and complies with the toggle count. The RL agent determines time slots and selects one of the time slots. Hybrid setpoints are generated to control the water distribution network within the selected time slot.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for computer aided design, including obtaining, by a shape modeling computer program, a selection of first geometry defined in a data structure used by the shape modeling computer program to represent a three-dimensional model of an object and an indication of an amount of desired prismatification, wherein the first geometry is defined in the data structure using a control mesh for a smooth surface representation of the first geometry. The shape modeling computer program produces second geometry defined in the data structure based on the indication of the amount of desired prismatification, wherein the second geometry replaces the first geometry in representing the three-dimensional model of the object. The shape modeling computer program provides the three-dimensional model of the object, with the second geometry included in the three-dimensional model.

Abstract: A method and system provide the ability to build a game world. A story is obtained that provides a textual narrative of a sequence of events. Plot facilities and a set of constraints are extracted from the story. Each of the plot facilities is a conceptual location where an event happens in the story. Each constraint defines a spatial relation between plot facilities. A map is generated based on the set of constraints by: generating a terrain of two dimensional (2D) polygons that is each associated with a biome type, and assigning each plot facility to a point on the terrain. The assigning complies with a maximum number of constraints and utilizes reinforcement learning (RL) to optimize positions of the points.

Abstract: A method and system provide for augmenting a photograph. An unlabeled photograph is obtained. A weakly augmented photograph and a strongly augmented photograph are obtained from the unlabeled photograph based on different types of data augmentation methods. The weakly augmented photograph is processed through a model to generate multiple weakly augmented photograph class predictions (with assigned probabilities). The multiple weakly augmented photograph class predictions are converted into positive pseudo-labels (indicating a presence of a class) or negative pseudo-labels (indicating absence of a class) using different fixed percentile thresholds. The strongly augmented photograph is processed through the model to generate a strongly augmented photograph class prediction. The model is trained to make the strongly augmented photograph label prediction match the positive pseudo-label via a cross-entropy loss. The trained model is then utilized to label the unlabeled photograph with multiple labels.

Abstract: A method and system provide the ability to process a construction domain query. A construction specification document that has textual content organized under multiple headings is obtained. The multiple headings in the textual content are identified. Each of the identified multiple headings is treated as a chunk and heading embeddings are generated and stored for each chunk. A user query is obtained. Based on the stored heading embeddings, the user query is performed within each of the chunks to retrieve relevant textual content from each chunk. The relevant textual content from each of the chunks is consolidated into a query response.

Abstract: A method and apparatus for automatically presenting a dimension of a depiction of an object is disclosed. One embodiment is evidenced by a method which comprises accepting a hit point on the depiction of the object and automatically presenting the dimension according to the plurality of surface boundary edges. The surface of the depicted object may be determined by collecting a plurality of connected triangles and edges that form the surface, then for each connected triangle, determining the triangle edges that form the border of the surface, and the triangle border vertices, and finally collating the triangle edges to produce the plurality of surface boundary edges from the triangle border edges.

Abstract: In various embodiments, a frame system application generates a design of a frame system associated with a building. The frame system application determines potential frame locations based on a frame grid for a structural system and a computer-aided design of the structural system and then bifurcates the potential frame locations based on a building load centroid to generate frame groups. Based on the frame groups, the frame system application generates a genetic algorithm that determines values for location counts associated with the frame groups based on an objective function that quantifies design objective(s). The frame system application executes the genetic algorithm on a value for the objective function that is associated with first values for the location counts to determine second values for the location counts. Based on the frame groups and the second values for the location counts, the frame system application generates the design of the frame system.

Abstract: A computer-implemented method of generating one or more variable stiffness structures includes determining a thickness of a first portion of a variable stiffness structure; determining a pressure that is to be applied to a surface of the first portion; selecting a first predetermined value for a stiffness attribute based on the thickness of the first portion and the pressure; and generating a model of at least part of the variable stiffness structure that includes the first portion, wherein the first portion has the predetermined value for the stiffness attribute.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, include: obtaining one or more design criteria for a modeled object; iteratively modifying a three-dimensional shape of the modeled object in accordance with the one or more design criteria, determining layer boundaries between three or more discrete layers for the three-dimensional shape based on differences among multiple milling depths identified for respective milling lines in a density-based representation of the three-dimensional shape, including adjusting the density-based representation of the three-dimensional shape to reassign milling depths for at least a portion of the milling lines such that the milling depths for the milling lines correspond to the layer boundaries between the three or more discrete layers for the three-dimensional shape, thereby changing the three-dimensional shape of the modeled object; and providing the three-dimensional shape of the modeled object for use in manufacturing a physical structur

Abstract: One embodiment of the present invention sets forth a technique for performing 3D shape generation. This technique includes generating semantic features associated with an input sketch. The technique also includes generating, using a generative machine learning model, a plurality of predicted shape embeddings from a set of fully masked shape embeddings based on the semantic features associated with the input sketch. The technique further includes converting the predicted shape embeddings into one or more 3D shapes. The input sketch may be a casual doodle, a professional illustration, or a 2D CAD software rendering. Each of the one or more 3D shapes may be a voxel representation, an implicit representation, or a 3D CAD software representation.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for detecting self-intersections of a surface mesh, include: obtaining (i) a surface mesh including vertices and edges that define triangles that approximate a surface of a three-dimensional model of a three-dimensional object, and (ii) a Delaunay diagram of the vertices of the surface mesh, wherein the Delaunay diagram includes a volumetric mesh including tetrahedrons having vertices at the vertices of the surface mesh; determining, using the Delaunay diagram, data indicating at least one intersection among the triangles of the surface mesh; and causing a refinement of the surface mesh based on the data indicating the at least one intersection, wherein the refinement removes the at least one intersection among the triangles of the surface mesh.

Abstract: A method and system provide the ability to extend immutable data. A first version of immutable design data is created. The first version of immutable design data is edited to create a second version of immutable design data. Subsequent to the creation of the second version of immutable data, the first version of immutable design data is augmented with first extendable data without creating a new version of immutable data.

Abstract: A method of modifying metadata associated with a media file is described the method comprising receiving a first media file; receiving a first metadata associated with the first media file; receiving an identifier of a second media file; and modifying the first metadata to include the identifier of the second media file.

Abstract: A method and system provide for annotating a computer-aided design (CAD) drawing. Existing drawings are obtained and include annotations and geometries that serve as hosts. A machine learning (ML) model is trained on the extracted geometries and annotations. A new drawing is obtained. First user input selecting a first geometry in the new drawing is received and the first annotation is created. The ML model generates potential new hosts and annotations. The potential new annotations are displayed in the new drawing and second user input selects one of the potential new annotations to utilize as one or more new annotations.

Abstract: A method and system provide the ability to interpolate fluids. At least two keyframes are produced, for a physics based fluid simulation. The keyframes are within a continuous-time framework and separated by a defined interval. Each keyframe includes one or more fluid elements having a corresponding state. Data is prepared utilizing a pre-trained transformer-based network by: (i) handling a tokenization process in a physics-adapted context; and (ii) generating temporal embeddings for states of the one or more fluid elements. Based on the prepared data, a time-continuous density is prepared for substeps between the two keyframes using a density network.

Abstract: In various embodiments a computer-implemented method for providing sustainability insights to a user designing an object. The method includes determining a value of a sustainability metric associated with a design of an object; determining a target value for the sustainability metric; determining an amount of progress made towards achieving the target value for the sustainability metric based on the value of the sustainability metric and the target value for the sustainability metric; and displaying, via a graphical user interface (GUI), a visual indication of the amount of progress made towards achieving the target value for the sustainability metric.

Abstract: In various embodiments, a knowledge capture application automatically captures knowledge associated with computer-based tasks. While a computer-based task is performed during a computer-based session, the knowledge capture application performs operation(s) on a user utterance to automatically generate a knowledge item. The knowledge capture application performs classification operation(s) on the knowledge item to generate a categorized knowledge item that associates the knowledge item with a first category. Subsequently, the knowledge capture application modifies a first element included in a graphical user interface to visually indicate an actual capture level associated with the first category. The knowledge capture application generates a knowledge dataset based on the first categorized knowledge item for storage or display.

Abstract: A method and system provide for augmenting a photograph. An unlabeled photograph is obtained. A weakly augmented photograph and a strongly augmented photograph are obtained from the unlabeled photograph based on different types of data augmentation methods. The weakly augmented photograph is processed through a model to generate multiple weakly augmented photograph class predictions (with assigned probabilities). The multiple weakly augmented photograph class predictions are converted into positive pseudo-labels (indicating a presence of a class) or negative pseudo-labels (indicating absence of a class) using different fixed percentile thresholds. The strongly augmented photograph is processed through the model to generate a strongly augmented photograph class prediction. The model is trained to make the strongly augmented photograph label prediction match the positive pseudo-label via a cross-entropy loss. The trained model is then utilized to label the unlabeled photograph with multiple labels.

Abstract: In various embodiments, a training application trains machine learning models to perform tasks associated with 3D CAD objects that are represented using B-reps. In operation, the training application computes a preliminary result via a machine learning model based on a representation of a 3D CAD object that includes a graph and multiple 2D UV-grids. Based on the preliminary result, the training application performs one or more operations to determine that the machine learning model has not been trained to perform a first task. The training application updates at least one parameter of a graph neural network included in the machine learning model based on the preliminary result to generate a modified machine learning model. The training application performs one or more operations to determine that the modified machine learning model has been trained to perform the first task.

Abstract: A computer-implemented method for generating a virtual collection of digital materials includes: generating a virtual three-dimensional (3D) design workspace; and, in response to a first operation that is associated with a first media file: automatically generating a virtual 3D representation of the first media file; and automatically incorporating the virtual 3D representation of the first media file into the virtual 3D design workspace.