Abstract: In various examples, updating model cards with information associated with modifications to machine learning models are described herein. For instance, systems and methods described herein may automatically update model cards with supplemental information about modifications that occur to models. To update a model card, the unique identifier may be used to identify that a modification occurred to a model—such as by performing fine-tuning, optimization, quantization, and/or any other modifying technique—in order to generate a new version of the model. Supplemental information about the modification is then obtained, such as by decoding the unique identifier, testing the new version of the model, and/or using data describing how the model was modified. The model card may then automatically be updated with the supplemental information. Additionally, this process may continue to repeat as additional modifications occur to the model to generate new versions of the model.

Abstract: A V2X computing system to manage air quality by reducing carbon emissions by changing a vehicle mode of operation from fuel to battery. The system identifies the position of the vehicle and changes the mode of the vehicle to a target battery-based vehicle. The system calculates how much aggregated emission can be allowed within a defined time range around a geo-fencing area to identify appropriate distribution of battery-operated vehicle and fossil fuel-based vehicle to ensure the required air-quality. The system leverages historical data around environmental parameters like wind direction, speed, etc. for a period to calculate the emission requirements vs. predicted emission. The system uses this data to dynamically switch vehicle from using fuel power to battery power based on the capability available within the vehicle for environmental sustainability. The system further spaces vehicles appropriately to ensure the air quality is maintained on the road in a geo-fenced area.

Abstract: In various examples, techniques for automatically generating and maintaining data cards for datasets is described herein. Systems and methods are disclosed that process a dataset in order to identify relevant information associated with the dataset. For example, the dataset may include and/or be associated with sources of information—such as files, documents, links, memos, research papers, annotations, labels, and/or the like—that describe data instances (e.g., images, audio clips, point clouds, etc.) included in the dataset. These sources of information may then be analyzed to retrieve the relevant information associated with the dataset. Systems and methods are then further disclosed that may use one or more language models to process input data associated with the relevant information in order to generate a data card associated with the dataset.

Abstract: One or more computer processors identifying at least one collaborating user within a virtual collaborative environment. The one or more computer processors define at least one volumetric boundary within the virtual collaborative environment. The one or more computer processors generate media content initiated by the at least one identified collaborating user within the at least one defined volumetric boundary. The one or more computer processors selectively record the generated media content within the at least one defined volumetric boundary.

Abstract: A method for device and role-based authentication includes generating a near-body electrical-field via a communication transmitter associated with a user, where a plurality of IoT devices is further associated with the user attempting authentication for access to a location. The method further includes receiving the near-body electrical-field via a communication receiver on an authentication device for identifying the user attempting the authentication. In response to identifying the user based on the near-body electrical-field, the method further includes analyzing a plurality of activities associated with a role for the user and analyzing plurality of resources associated with the user, wherein the plurality resources include at least the plurality of IoT devices associated with the user.

Abstract: In various examples, techniques for automatically generating and maintaining data cards for datasets is described herein. Systems and methods are disclosed that process a dataset in order to identify relevant information associated with the dataset. For example, the dataset may include and/or be associated with sources of information-such as files, documents, links, memos, research papers, annotations, labels, and/or the like-that describe data instances (e.g., images, audio clips, point clouds, etc.) included in the dataset. These sources of information may then be analyzed to retrieve the relevant information associated with the dataset. Systems and methods are then further disclosed that may use one or more language models to process input data associated with the relevant information in order to generate a data card associated with the dataset.

Abstract: An embodiment for controlling material movement with swarm power generating robots in a multi-machine environment is provided. The embodiment may include receiving data relating to an activity and one or more material handling devices to perform the activity. The embodiment may also include identifying one or more characteristics of one or more objects associated with the activity. The embodiment may further include predicting an amount of power required to transport the one or more objects. The embodiment may also include in response to determining at least one material handling device is unable to produce the required amount of power, identifying one or more power generation robots capable of transmitting the required amount of power to the at least one material handling device. The embodiment may further include deploying the one or more power generation robots to a target location of the at least one material handling device.

Abstract: In various examples, automatic generation of model cards for machine learning models is described herein. Systems and methods are disclosed that use one or more language models, which process input data representing information associated with a model (e.g., a machine learning model, an AI model, a neural network, etc.), to automatically generate a model card to associate with the model. As described herein, the information associated with the model may include at least a portion of source code used to generate the model, one or more documents that describe the model, one or more previously generated model cards, and/or any other information associated with the model. Additionally, in some examples, additional data may be input into the language model(s) to generate the model card, such as data representing questions for retrieving relevant information and/or data representing reference information associated with one or more other models.

Abstract: In various examples, bias detection and mitigation for machine learning models is described herein. Systems and methods are disclosed that train or otherwise update one or more machine learning models (e.g., model(s)) using a dataset that includes images of people and annotations indicating skin tones associated with the people as depicted by the images. In some examples, images may be associated with various groups, where each group is associated with a range of lighting values and a range of hue values associated with the skin tones. After an initial training the model(s), the systems and methods may then evaluate the model(s) in order to identify groups for which bias exist and mitigate the bias using further training.

Abstract: According to a technique of designing an intelligent workflow, a processor determines an intelligent workflow including a plurality of workflow steps to be performed. The processor performs digital twin simulation of performance of the intelligent workflow in a physical production environment. The processor determines, based on the digital twin simulation, at least one type of intelligence among a plurality of types of intelligences to be allocated to each of multiple of the plurality of workflow steps. The processor determines multiple types of intelligences for the plurality of workflow steps. The processor allocates and deploys, in the physical production environment, production resources possessing the determined types of intelligences. The processor thereafter iteratively optimizes the intelligent workflow based on observation of execution of the intelligent workflow by the deployed performance resources in the physical production environment.

Abstract: Dynamic modification of an extended reality environment include receiving an input associated with a set of rules for rendering a first primary extended reality (XR) environment on a first user device. The first primary XR environment is rendered on the first user device associated with a first avatar and includes the first avatar and a second avatar. A first action of the second avatar of the first set of avatars is detected. Based on the detected first action, the execution of at least a first rule of the set of rules is triggered. The first primary XR environment is modified by masking or unmasking a portion of the second avatar within the first primary XR environment based on the execution of at least the first rule and the modified first primary XR environment is rendered on the first user device.

Abstract: Aggregating worker skills and wearable robotic system capabilities is provided. Details of an activity that a worker is preparing to perform utilizing a wearable robotic system are retrieved. Skills of the worker related to the activity are retrieved. Capabilities of the wearable robotic system related to the activity are retrieved. An analysis of the details of the activity that the worker is preparing to perform utilizing the wearable robotic system, the skills of the worker, and the capabilities of the wearable robotic system is performed. An aggregation of the skills of the worker and the capabilities of the wearable robotic system to perform the activity is determined based on the analysis. It is determined that the worker can perform the activity utilizing the wearable robotic system based on the aggregation of the skills of the worker and the capabilities of the wearable robotic system.

Abstract: Hardware and software on a computing device is analyzed based on a regulatory profile for the computing device and regulatory compliance for an entity associated with the computing device. A determination is made whether at least one of the hardware and software on the computing device includes at least one regulatory non-compliance issue. In response to determining that at least one of the hardware and software on the computing device includes at least one regulatory non-compliance issue, one or more scripts are executed on the hardware and software on the computing device to cause the hardware and software to resolve the at least one regulatory non-compliance issue based on the regulatory profile for the computing device.

Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: storing into a data repository internet of things (IoT) sensor data of a plurality of IoT devices disposed within a workflow environment that includes one or more physical asset; performing a simulation to simulate operating performance of the one or more physical asset disposed within the workflow environment, wherein the performing the simulation to simulate operating performance of the one or more physical asset disposed within the workflow environment includes using historical IoT data of the IoT sensor data; detecting, in dependence on the performing the simulation, that an alert condition is present in the workflow environment; and prompting one or more worker within the workflow environment to take action in response to the detecting that the alert condition is present in the workflow environment.

Abstract: In various examples, metadata associated with one or more models may be captured during a training process and stored in association with the model(s). The metadata may then be used, in some embodiments, for enforcing execution of the model(s). For instance, the model(s) may be trained during at least a portion of the training process. During at least a second portion of the training process, one or more attributes associated with the model(s) may be determined. The attribute(s) may then be stored as metadata in association with the model(s). Additionally, in some embodiments, an endpoint may request to execute the model(s). Responsive to the request, and based at least on evaluating the metadata with respect to one or more criteria associated with the endpoint, a determination may be made regarding whether or not to provide the model(s) to the endpoint for execution.

Abstract: Dynamic modification of an extended reality environment include receiving an input associated with a set of rules for rendering a first primary extended reality (XR) environment on a first user device. The first primary XR environment is rendered on the first user device associated with a first avatar and includes the first avatar and a second avatar. A first action of the second avatar of the first set of avatars is detected. Based on the detected first action, the execution of at least a first rule of the set of rules is triggered. The first primary XR environment is modified by masking or unmasking a portion of the second avatar within the first primary XR environment based on the execution of at least the first rule and the modified first primary XR environment is rendered on the first user device.

Abstract: One or more computer processors identifying at least one collaborating user within a virtual collaborative environment. The one or more computer processors define at least one volumetric boundary within the virtual collaborative environment. The one or more computer processors generate media content initiated by the at least one identified collaborating user within the at least one defined volumetric boundary. The one or more computer processors selectively record the generated media content within the at least one defined volumetric boundary.

Abstract: A V2X computing system to manage air quality by reducing carbon emissions by changing a vehicle mode of operation from fuel to battery. The system identifies the position of the vehicle and changes the mode of the vehicle to a target battery-based vehicle. The system calculates how much aggregated emission can be allowed within a defined time range around a geo-fencing area to identify appropriate distribution of battery-operated vehicle and fossil fuel-based vehicle to ensure the required air-quality. The system leverages historical data around environmental parameters like wind direction, speed, etc. for a period to calculate the emission requirements vs. predicted emission. The system uses this data to dynamically switch vehicle from using fuel power to battery power based on the capability available within the vehicle for environmental sustainability. The system further spaces vehicles appropriately to ensure the air quality is maintained on the road in a geo-fenced area.

Abstract: An embodiment for controlling material movement with swarm power generating robots in a multi-machine environment is provided. The embodiment may include receiving data relating to an activity and one or more material handling devices to perform the activity. The embodiment may also include identifying one or more characteristics of one or more objects associated with the activity. The embodiment may further include predicting an amount of power required to transport the one or more objects. The embodiment may also include in response to determining at least one material handling device is unable to produce the required amount of power, identifying one or more power generation robots capable of transmitting the required amount of power to the at least one material handling device. The embodiment may further include deploying the one or more power generation robots to a target location of the at least one material handling device.

Abstract: Aspects of the present disclosure relate to extended reality (XR) device thermal load management. Thermal data associated with an XR device can be received. A determination can be made that a condition is met for offloading a thermal load of the XR device. One or more edge processing devices within the environment of the XR device can be identified. At least one edge processing device with sufficient computing resources for balancing the thermal load of the XR device can be selected. The XR device can be instructed to offload the thermal load to the selected at least one edge processing device.