Abstract: User-level privacy preservation is implemented within federated machine learning. An aggregation server may distribute a machine learning model to multiple users each including respective private datasets. Individual users may train the model using the local, private dataset to generate one or more parameter updates. Prior to sending the generated parameter updates to the aggregation server for incorporation into the machine learning model, a user may modify the parameter updates by applying respective noise values to individual ones of the parameter updates to ensure differential privacy for the dataset private to the user. The aggregation server may then receive the respective modified parameter updates from the multiple users and aggregate the updates into a single set of parameter updates to update the machine learning model. The federated machine learning may further include iteratively performing said sending, training, modifying, receiving, aggregating and updating steps.

Abstract: User-level privacy preservation is implemented within federated machine learning. An aggregation server may distribute a machine learning model to multiple users each including respective private datasets. Individual users may train the model using the local, private dataset to generate one or more parameter updates. Prior to sending the generated parameter updates to the aggregation server for incorporation into the machine learning model, a user may modify the parameter updates by applying respective noise values to individual ones of the parameter updates to ensure differential privacy for the dataset private to the user. The aggregation server may then receive the respective modified parameter updates from the multiple users and aggregate the updates into a single set of parameter updates to update the machine learning model. The federated machine learning may further include iteratively performing said sending, training, modifying, receiving, aggregating and updating steps.

Abstract: The present disclosure relates to microbial compositions for enhancing bile acid metabolism in a subject, methods of manufacturing the compositions, and methods of selecting subjects in need of treatment with the compositions. Certain bile acid transforming or converting genes and bile acid transforming or converting bacteria are used to select desired bacterial isolates or suitable donors or patients, predict treatment success, or track treatment response.

Abstract: User-level privacy preservation is implemented within federated machine learning. An aggregation server may distribute a machine learning model to multiple users each including respective private datasets. Individual users may train the model using the local, private dataset to generate one or more parameter updates. Prior to sending the generated parameter updates to the aggregation server for incorporation into the machine learning model, a user may modify the parameter updates by applying respective noise values to individual ones of the parameter updates to ensure differential privacy for the dataset private to the user. The aggregation server may then receive the respective modified parameter updates from the multiple users and aggregate the updates into a single set of parameter updates to update the machine learning model. The federated machine learning may further include iteratively performing said sending, training, modifying, receiving, aggregating and updating steps.

Abstract: Systems and methods for platform occupancy detection on an aerial device are described. An aerial platform may be provided for the aerial device. A worker may perform operations in the aerial platform and connect to an interlock system thereof. The worker may have an associated transmitter configured to transmit a signal. A first receiver and a second receiver may be disposed on the aerial platform to receive the signal. Based on the time difference between receiving the signal at the first receiver and receiving the signal at the second receiver, the position of the transmitter may be triangulated. If the transmitter is determined to be within the aerial platform and the worker is not connected to the interlock system, preventive actions may be taken before operation of the aerial platform is permitted.

Abstract: Herein are techniques for domain adaptation of a machine learning (ML) model. These techniques impose differential privacy onto federated learning by the ML model. In an embodiment, each of many client devices receive, from a server, coefficients of a general ML model. For respective new data point(s), each client device operates as follows. Based on the new data point(s), a respective private ML model is trained. Based on the new data point(s), respective gradients are calculated for the coefficients of the general ML model. Random noise is added to the gradients to generate respective noisy gradients. A combined inference may be generated based on: the private ML model, the general ML model, and one of the new data point(s). The noisy gradients are sent to the server. The server adjusts the general ML model based on the noisy gradients from the client devices. This client/server process may be repeated indefinitely.

Abstract: Systems and methods are disclosed to improve a topic modeling system that tunes a topic model for a set of topics from a corpus of documents, by allowing users to pre-inform the tuning process with bias parameters for desired associations in the topic model. In embodiments, the topic model may be a Latent Dirichlet Allocation (LDA) model. In embodiments, the bias parameter may indicate a fixed association where a particular word in a particular document is associated with a particular topic. In embodiments, the bias parameter may specify a weight value that biases the inference process with regard to a particular association. Advantageously, the disclosed features allow users to specify a small number of parameters to steer the tuning process towards a set of desired topics. As a result, the topic model may be generated more quickly and with more useful topics.

Abstract: Herein are techniques for domain adaptation of a machine learning (ML) model. These techniques impose differential privacy onto federated learning by the ML model. In an embodiment, each of many client devices receive, from a server, coefficients of a general ML model. For respective new data point(s), each client device operates as follows. Based on the new data point(s), a respective private ML model is trained. Based on the new data point(s), respective gradients are calculated for the coefficients of the general ML model. Random noise is added to the gradients to generate respective noisy gradients. A combined inference may be generated based on: the private ML model, the general ML model, and one of the new data point(s). The noisy gradients are sent to the server. The server adjusts the general ML model based on the noisy gradients from the client devices. This client/server process may be repeated indefinitely.

Abstract: A method for leashing an animal is provided, including steps of attaching a quick-release mechanism to a collar on an animal, and encircling the animal's abdomen by passing a handle end of the leash through an intermediate loop such that the leash forms a constrictive loop around the animal's abdomen. The leash includes a collar end formed from a loop of a first flexible web that captures the quick-release mechanism, and a handle end formed from a loop of a second flexible web. The intermediate loop is between the collar end and the handle end and formed of a third flexible web. The first flexible web overlaps the third flexible web and the overlapping portions are bonded together at a collar end of the intermediate loop. The second flexible web overlaps the third flexible web and the overlapping portions are bonded together at a handle end of the intermediate loop.

Abstract: Systems and methods are disclosed to improve a topic modeling system that tunes a topic model for a set of topics from a corpus of documents, by allowing users to pre-inform the tuning process with bias parameters for desired associations in the topic model. In embodiments, the topic model may be a Latent Dirichlet Allocation (LDA) model. In embodiments, the bias parameter may indicate a fixed association where a particular word in a particular document is associated with a particular topic. In embodiments, the bias parameter may specify a weight value that biases the inference process with regard to a particular association. Advantageously, the disclosed features allow users to specify a small number of parameters to steer the tuning process towards a set of desired topics. As a result, the topic model may be generated more quickly and with more useful topics.

Abstract: A leash for an animal is provided that includes a collar end and a handle end positioned opposite the collar end. An intermediate portion of the leash is located between the collar end and the handle end of the leash. The intermediate portion includes two separate strands wherein a first end of each strand is attached at the collar end of the leash and a second end of each strand is attached at the handle end of the leash. The leash is configured to encircle the abdomen of the animal by passing the handle end of the leash through a loop formed between the two separate strands such that the leash forms a constrictive loop around the animal's midsection or abdomen, specifically the portion of the animal's torso located between animal's hind legs and the animal's ribs. Methods for using the leash are also disclosed.

Abstract: The fluid-segmentation device according to an embodiment of the present invention includes: a first conduit in which a first fluid flows, and a second conduit in which a second fluid immiscible with the first fluid flows. The second conduit of the fluid-segmentation device includes an intersection region, to which the first conduit is connected and into which the first fluid is introduced, and a first region downstream of the intersection region. The cross-sectional area of the intersection region of the second conduit in a plane perpendicular to the flow direction of the second fluid is less than the cross-sectional area of the first region of the second conduit in a plane perpendicular to the flow direction of the second fluid.

Abstract: Examples are disclosed that relate to an ultrafiltration system for quantum-dot (QD) purification. The ultrafiltration system comprises a pump having a low-pressure side and a high-pressure side, a size-exclusion membrane having a low-pressure side and a high-pressure side, and an inlet/outlet arrangement. An inlet arranged on the high-pressure side of the size-exclusion membrane is coupled fluidically to the high-pressure side of the pump. A product-enriched outlet is arranged on the high-pressure side of the size-exclusion membrane, fluidically downstream of the inlet. A product-depleted outlet is arranged on the low-pressure side of the size-exclusion membrane.

Abstract: A leash for an animal is provided that includes a collar end and a handle end positioned opposite the collar end. An intermediate portion of the leash is located between the collar end and the handle end of the leash. The intermediate portion includes two separate strands wherein a first end of each strand is attached at the collar end of the leash and a second end of each strand is attached at the handle end of the leash. The leash is configured to encircle the abdomen of the animal by passing the handle end of the leash through a loop formed between the two separate strands such that the leash forms a constrictive loop around the animal's midsection or abdomen, specifically the portion of the animal's torso located between animal's hind legs and the animal's ribs. Methods for using the leash are also disclosed.

Abstract: The bezel assembly for data reception, for use with a bill validator in a financial transactional device, includes a bezel housing and a data reception assembly. The bezel housing includes a customer-facing front portion and a back plate connectable to the bill validator that is mounted within the transactional device cabinet. The front portion includes an insertion/dispensing slot for receiving currency and a projecting hollow tongue forward of the casing. The forward-extending hollow tongue accommodates at least a portion of the data reception assembly. The bezel assembly can include a wireless communication function that is communicably connectable with a mobile device via a wireless communication method, a manual entry function, a biometric reader, one or more cameras for scanning and decrypting 2D barcodes and the like, thus enhancing the overall functionality of the financial transactional device.

Abstract: The bezel assembly for data reception, for use with a bill validator in a financial transactional device, includes a bezel housing and a data reception assembly. The bezel housing includes a customer-facing front portion and a back plate connectable to the bill validator that is mounted within the transactional device cabinet. The front portion includes an insertion/dispensing slot for receiving currency and a projecting protrusion forward of the casing. The forward-extending protrusion accommodates at least a portion of the data reception assembly. The bezel assembly can include a wireless communication function that is communicably connectable with a mobile device via a wireless communication method, a manual entry function, a biometric reader, one or more cameras for scanning and decrypting 2D barcodes and the like, thus enhancing the overall functionality of the financial transactional device.

Abstract: The bezel assembly for data reception, for use with a bill validator in a financial transactional device, includes a bezel housing and a data reception assembly. The bezel housing includes a customer-facing front portion and a back plate connectable to the bill validator that is mounted within the transactional device cabinet. The front portion includes an insertion/dispensing slot for receiving currency and a projecting protrusion forward of the casing. The forward-extending protrusion accommodates at least a portion of the data reception assembly. The bezel assembly can include a wireless communication function that is communicably connectable with a mobile device via a wireless communication method, a manual entry function, a biometric reader, one or more cameras for scanning and decrypting 2D barcodes and the like, thus enhancing the overall functionality of the financial transactional device.

Abstract: The bezel assembly for data reception, for use with a bill validator in a financial transactional device, includes a bezel housing and a data reception assembly. The bezel housing includes a customer-facing front portion and a back plate connectable to the bill validator that is mounted within the transactional device cabinet. The front portion includes an insertion/dispensing slot for receiving currency and a projecting protrusion forward of the casing. The forward-extending protrusion accommodates at least a portion of the data reception assembly. The bezel assembly can include a wireless communication function that is communicably connectable with a mobile device via a wireless communication method, a manual entry function, a biometric reader, one or more cameras for scanning and decrypting 2D barcodes and the like, thus enhancing the overall functionality of the financial transactional device.

Abstract: A vehicle side fairing having a first fairing panel fixedly secured to the underside of a trailer box, and a second fairing panel slidable mounted to plural cylindrical bars mounted on the underside of the trailer box in a fore and aft orientation. The second fairing panel is secured to the wheel set with a mechanical fastener that selectively joins and releases the second fairing panel and the wheel set.

Abstract: A non-synthetic, hydrophilic, biodegradable, biocompatible polysaccharide based non-toxic anti-adhesion hydrogel barrier is disclosed herein. The barrier of the present invention is formed by constructing a unique interpenetrating, crosslinked network with a unique porosity. Furthermore, the barrier of the present invention is comprised of tunable biopolymers for controllable mechanical robustness and degradation. The barrier of the present invention effectively reduces unwanted adhesions using non-synthetic components.