Abstract: The disclosure herein describes training a global model based on a plurality of data sets. The global model is applied to each data set of the plurality of data sets and a plurality of gradients is generated based on that application. At least one gradient quality metric is determined for each gradient of the plurality of gradients. Based on the determined gradient quality metrics of the plurality of gradients, a plurality of weight factors is calculated. The plurality of gradients is transformed into a plurality of weighted gradients based on the calculated plurality of weight factors and a global gradient is generated based on the plurality of weighted gradients. The global model is updated based on the global gradient, wherein the updated global model, when applied to a data set, performs a task based on the data set and provides model output based on performing the task.

Abstract: A battery pack includes battery modules each having a battery cell and a battery case housing the battery cell therein. The battery modules are stacked in a stacking direction so that a cooling passage for allowing a cooling medium to flow is defined between opposed surfaces of adjacent battery cases. One of the opposed surfaces has first ribs projecting toward the other of the opposed surfaces and extending parallel to each other along the one. The other of the opposed surfaces has second ribs projecting toward the one of the opposed surfaces and extending parallel to each other along the other. In the cooling passage, the first ribs and the second ribs intersect each other and end portions of the first ribs and end portions of the second ribs are in contact with each other in the stacking direction.

Abstract: A method of determining the consistency of training data for a machine translation system is disclosed. The method includes receiving a signal indicative of a source language corpus and a target language corpus. A textual string is extracted from the source language corpus. The textual string is aligned with the target language corpus to identify a translation for the textual string from the target language corpus. A consistency index is calculated based on a relationship between the textual string from the source language corpus and the translation. An indication of the consistency index is stored on a tangible medium.

Abstract: A battery pack includes battery modules each having a battery cell and a battery case housing the battery cell therein. The battery modules are stacked in a stacking direction so that a cooling passage for allowing a cooling medium to flow is defined between opposed surfaces of adjacent battery cases. One of the opposed surfaces has first ribs projecting toward the other of the opposed surfaces and extending parallel to each other along the one. The other of the opposed surfaces has second ribs projecting toward the one of the opposed surfaces and extending parallel to each other along the other. In the cooling passage, the first ribs and the second ribs intersect each other and end portions of the first ribs and end portions of the second ribs are in contact with each other in the stacking direction.

Abstract: A system described herein includes a receiver component that receives an output translation from a machine translation system, wherein the output translation is in a target language and is based at least in part upon an input to the machine translation system in a source language, and wherein the input to the machine translation system includes a first term in the source language and the output translation includes a second term in the target language that corresponds to the first term. The system additionally includes a replacer component in communication with the receiver component that accesses a dictionary of term correspondences, wherein the dictionary of term correspondences includes an indication that the input first term in the source language is desirably translated to a third term in the target language, and wherein the replacer component is configured to automatically replace the second term with the third term to modify the output translation.

Abstract: A method of determining the consistency of training data for a machine translation system is disclosed. The method includes receiving a signal indicative of a source language corpus and a target language corpus. A textual string is extracted from the source language corpus. The textual string is aligned with the target language corpus to identify a translation for the textual string from the target language corpus. A consistency index is calculated based on a relationship between the textual string from the source language corpus and the translation. An indication of the consistency index is stored on a tangible medium.