Abstract: According to some aspects, techniques are provided for building and deploying a machine learning application that do not require a user to have expert knowledge of machine learning or programming. These techniques may be executed by a system that provides a graphical user interface which allows a user to visually define a workflow for a machine learning application, without requiring the user to be an expert in machine learning. The system may automatically represent the workflow as a specification that may be used to build and deploy a machine learning application. The system may automatically execute the workflow in a series of stages while managing data flow and execution context between the stages. Such an execution process may provide flexibility in execution so that a user can build a complex machine learning application without it being necessary for the user to have detailed knowledge of how execution is managed.

Abstract: A data management system (DMS) may support intelligent snapshot protection techniques. For example, the DMS may backup a computing system and facilitate that capture and storage of snapshots of the computing system. The DMS may identify a deletion event associated with one or more of the snapshots, the deletion event being associated with the one or more snapshots being deleted at a first time. The DMS may determine that deletion event is anomalous and retain the one or more snapshots beyond the first time. For example, the DMS may determine that one or more parameters associated with the deletion event are indicative of the deletion event being anomalous. In response, the DMS may retain the one or more snapshots for a retention period beyond the first time.

Abstract: In some examples, a wafer bow measurement system comprises a measurement unit including: a wafer support assembly to impart rotational movement to a measured wafer supported in the measurement unit; an optical sensor; a calibration standard to calibrate the optical sensor; a linear stage actuator to impart linear direction of movement to the optical sensor; a wafer centering sensor to determine a centering of the measured wafer supported in the measurement unit; and a wafer alignment sensor to determine an alignment of the measured wafer supported in the measurement unit.

Abstract: Provided are methods of assigning a LAG+, LAG?, or PRO immunotype to a cancer patient based on the frequencies of LAG-3+CD8+T-cells, Ki67+CD8+T-cells, Tim-3+CD8+T-cells, and ICOS+CD8+T-cells in a peripheral blood sample from the patient, and selecting an anti-cancer therapy, for example, an immune checkpoint blockade (ICB) therapy, based on the patient's immunotype.

Abstract: According to some aspects, techniques are provided for building and deploying a machine learning application that do not require a user to have expert knowledge of machine learning or programming. These techniques may be executed by a system that provides a graphical user interface which allows a user to visually define a workflow for a machine learning application, without requiring the user to be an expert in machine learning. The system may automatically represent the workflow as a specification that may be used to build and deploy a machine learning application. The system may automatically execute the workflow in a series of stages while managing data flow and execution context between the stages. Such an execution process may provide flexibility in execution so that a user can build a complex machine learning application without it being necessary for the user to have detailed knowledge of how execution is managed.

Abstract: Techniques are generally described for fusing sensor data of different modalities using a transformer. In various examples, first sensor data may be received from a first sensor and second sensor data may be received from a second sensor. A first feature representation of the first sensor data may be generated using a first machine learning model and a second feature representation of the second sensor data may be generated using a second machine learning model. In some examples, a modified first feature representation of the first sensor data may be generated based at least in part on a self-attention mechanism of a transformer encoder. The modified first feature representation may be generated based at least in part on the first feature representation and the second feature representation. A computer vision task may be performed using the modified first feature representation.

Abstract: A valve shaft seal for a valve shaft of an internal combustion engine, includes at least one sealing lip of an elastomer material having a first inside diameter and configured to surround the valve shaft and a plurality of projections distributed uniformly around a circumference of the sealing lip and directed toward the valve shaft, a second inside diameter, smaller than the first diameter, being formed by the projections.