Abstract: Aspects of the present disclosure provide systems, methods, and computer-readable storage media that support generative artificial intelligence (AI)-assisted analytics of structured data sets. For example, a system may receive a prompt that includes information associated with a structured data set which includes at least some numerical data. The system may provide the prompt as input to an agent orchestrator to select one or more generative AI agents to perform analytics tasks corresponding to the information. The agent orchestrator includes a trained AI classifier configured to select the one or more generative AI agents from a plurality of generative AI agents. The system may execute an ensemble model to generate a response to the prompt based on the structured data set. The ensemble model includes the one or more generative AI agents. The system may output a graphical user interface (GUI) that includes one or more elements based on the response.

Abstract: Systems and methods for responsible AI compliance and governance management in AI Products are disclosed. The system receives a request to assess an enterprise product associated with a specific application. Further, the system may determine a plurality of datasets associated with the AI model of the enterprise product. Furthermore, the system generates a training dataset and a test dataset for the determined plurality of datasets associated with the AI model. The system generates a ranked list of recommended metrics for the enterprise product based on the generated training dataset and the test dataset. The system further determines a mitigation strategy for the enterprise product based on the generated ranked list of recommended metrics. Furthermore, the system creates a feedback loop for continuous training and tuning the AI model and the plurality of datasets based on the determined mitigation strategy.

Abstract: In some examples, structure-based multi-intent email classification may include receiving an email thread that includes a plurality of emails including an email that is to be classified, and identifying, for the email thread, a process associated with the email thread, Based on the process associated with the email thread and for each sentence of the email that includes a plurality of sentences, a corresponding classifier may be determined from a plurality of classifiers, and applied to each sentence of the email to generate classified sentences. At least one entity may be extracted from each of the classified sentences of the email. Based on the at least one extracted entity, the email may be classified.

Abstract: In some examples, structure-based multi-intent email classification may include receiving an email thread that includes a plurality of emails including an email that is to be classified, and identifying, for the email thread, a process associated with the email thread, Based on the process associated with the email thread and for each sentence of the email that includes a plurality of sentences, a corresponding classifier may be determined from a plurality of classifiers, and applied to each sentence of the email to generate classified sentences. At least one entity may be extracted from each of the classified sentences of the email. Based on the at least one extracted entity, the email may be classified.

Abstract: Systems and methods provide for improving the accuracy of a location system. The location system can capture partial phase vector data from one or more access points (APs). The location system can capture associated data associated with the partial phase vector data across multiple dimensions, such as identity data of the APs and client devices generating the partial phase vector data and frequency band data, location data, a time and date, and other data associated with the partial phase vector data. The location system can determine correlation data across the multiple dimensions using the first partial phase vector data and the associated data. The location system can a cause of the partial phase vector data based on the correlation data. The location system can perform one or more remediation actions based on the cause of the partial phase vector data.

Abstract: Offloading of location computation from a location server to an access point through the use of projections on base phase vectors may be provided. First, an Access Point (AP) may receive a set of two or more base phase vectors from a location server. Next, the AP may measure a measured phase vector for a first signal from a user device. Then, the AP can determine projection values based on a comparison of the measured phase vector to each base phase vector. From these comparisons, the AP can determine a subset of base phase vectors with the highest projection values. The AP can then send the projection values and the subset of base phase vectors to the location server, wherein the location server determines the device location from these projection values and subset of base phase vectors.

Abstract: A model optimization system monitors a model deployed to an external system to determine the performance of the model and to replace the model with one of a plurality of models stored to a model repository if degradation of model performance is detected or if one of the models in the plurality of models is evaluated as having better performance than the model deploy the external system. A model evaluation trigger can be generated based on dates or data criteria. Various metrics are used in the model evaluation to calculate values of a model optimization function for each of the plurality of models. If a model that is better optimized than the deployed model is identified from the model evaluation, then the deployed model is replaced with the identified model and the external system continues to use the deployed model.

Abstract: Location-based, context-aware challenge-response authentication may be provided. First, a challenge may be provided to a user. The challenge may be based on a context corresponding to the user. The context corresponding to the user may comprise a location of a device associated with the user within an environment. Next, in response to providing the challenge, a response to the challenge may be received from the user. Then, in response to receiving the response to the challenge, it may be determined that the response is a correct answer to the challenge. In response to determining the response is the correct answer, a privilege may be provided to the user.

Abstract: Determining a device's location in a space in real time is computing intensive. To offload some of the workload in conducting this hyperlocation, the access points in the network conduct some of process in determining the location of a device. The cloud determines a restricted AoA search area based on previous client locations. After this determination, a three-dimensional (3D) AoA search is conducted by each AP in the restricted area (restricted by a range of azimuth directions) for a device. Finally, each AP reports a location(s) for the device, which comprises weights for selected angular sectors. The cloud can then construct a probability heat map for location computation from the weights provided from each AP for the device.

Abstract: In one embodiment, a service obtains spatial information regarding a physical area. The service estimates locations of a device within the physical area over time, based on wireless signals sent by the device. The service generates a set of images based on the spatial information regarding the physical area and on the estimated locations of the device within the physical area over time. The service updates an estimated location of the device by inputting the generated set of images to a machine learning model trained to minimize a location estimation error.

Abstract: In one embodiment, a service obtains spatial information regarding a physical area. The service estimates locations of a device within the physical area over time, based on wireless signals sent by the device. The service generates a set of images based on the spatial information regarding the physical area and on the estimated locations of the device within the physical area over time. The service updates an estimated location of the device by inputting the generated set of images to a machine learning model trained to minimize a location estimation error.

Abstract: In one embodiment, a traffic analysis service receives payload data from packets sent by a sensor tag in a network. The service forms a payload signature for the sensor tag, based on the payload data. The payload signature is indicative of one or more bytes in the payload that vary across the packets. The service identifies a portion of the payload data as potentially including a sensor measurement, based on the payload signature. The service uses a machine learning classifier to assign a sensor measurement type to the identified portion of the payload data.

Abstract: In one embodiment, a traffic analysis service receives payload data from packets sent by a sensor tag in a network. The service forms a payload signature for the sensor tag, based on the payload data. The payload signature is indicative of one or more bytes in the payload that vary across the packets. The service identifies a portion of the payload data as potentially including a sensor measurement, based on the payload signature. The service uses a machine learning classifier to assign a sensor measurement type to the identified portion of the payload data.

Abstract: An enterprise system configures access point devices at an enterprise location to communicate with a location determination system. The location determination system receives wireless signal attributes of user computing devices broadcasting Wi-Fi signal data at the enterprise location from one or more access point devices. For a particular time window, the location determination system determines aggregated features of received wireless signal data across all access point devices, and classifies each of the user computing devices as moving or stationary by applying the wireless signal data to a model. For each of the user computing devices determined to be moving, the location determination system calculates a respective position of the user computing device based on the wireless signal data. For each of the user computing devices determined to be stationary, the location determination system does not calculate a respective position of the respective user computing device.

Abstract: A plurality of access points located within a building structure may provide various location-based services, such as navigation, to mobile devices within the building structure. To more accurately determine the location of the plurality of access points, a floorplan of the interior of the building structure may be obtained. A plurality of images of the interior of the building structure is then obtained. At least one of the images includes an image of at least one of the plurality of access points. A plurality of reference points may also be located within the building structure and obtained. Based on the plurality of images, reference points, and the floorplan, a composite image of the interior of the structure is generated. Based on the composite image, the location of each of the plurality of access points is determined.

Abstract: Offloading of location computation from a location server to an access point through the use of projections on base phase vectors may be provided. First, an Access Point (AP) may receive a set of two or more base phase vectors from a location server. Next, the AP may measure a measured phase vector for a first signal from a user device. Then, the AP can determine projection values based on a comparison of the measured phase vector to each base phase vector. From these comparisons, the AP can determine a subset of base phase vectors with the highest projection values. The AP can then send the projection values and the subset of base phase vectors to the location server, wherein the location server determines the device location from these projection values and subset of base phase vectors.

Abstract: Offloading of location computation from a location server to an access point through the use of projections on base phase vectors may be provided. First, an Access Point (AP) may receive a set of two or more base phase vectors from a location server. Next, the AP may measure a measured phase vector for a first signal from a user device. Then, the AP can determine projection values based on a comparison of the measured phase vector to each base phase vector. From these comparisons, the AP can determine a subset of base phase vectors with the highest projection values. The AP can then send the projection values and the subset of base phase vectors to the location server, wherein the location server determines the device location from these projection values and subset of base phase vectors.

Abstract: Determining a device's location in a space in real time is computing intensive. To offload some of the workload in conducting this hyperlocation, the access points in the network conduct some of process in determining the location of a device. The cloud determines a restricted AoA search area based on previous client locations. After this determination, a three-dimensional (3D) AoA search is conducted by each AP in the restricted area (restricted by a range of azimuth directions) for a device. Finally, each AP reports a location(s) for the device, which comprises weights for selected angular sectors. The cloud can then construct a probability heat map for location computation from the weights provided from each AP for the device.

Abstract: Offloading of location computation from a location server to an access point through the use of projections on base phase vectors may be provided. First, an Access Point (AP) may receive a set of two or more base phase vectors from a location server. Next, the AP may measure a measured phase vector for a first signal from a user device. Then, the AP can determine projection values based on a comparison of the measured phase vector to each base phase vector. From these comparisons, the AP can determine a subset of base phase vectors with the highest projection values. The AP can then send the projection values and the subset of base phase vectors to the location server, wherein the location server determines the device location from these projection values and subset of base phase vectors.

Abstract: Location-based, context-aware challenge-response authentication may be provided. First, a challenge may be provided to a user. The challenge may be based on a context corresponding to the user. The context corresponding to the user may comprise a location of a device associated with the user within an environment. Next, in response to providing the challenge, a response to the challenge may be received from the user. Then, in response to receiving the response to the challenge, it may be determined that the response is a correct answer to the challenge. In response to determining the response is the correct answer, a privilege may be provided to the user.