Abstract: A safety-oriented control unit and method for synchronizing measurement values delivered in an individual measurement phase, wherein the measurement values are processed by the safety-oriented controller in a monitoring phase in accordance with a safety function, where individual counter increments of the measurement values are compared with one another in comparison during a current monitoring cycle of the monitoring phase with a previous monitoring cycle, a maximum counter increment is determined, the measurement values that do not have the maximum counter increment are adjusted based on the maximum counter increment, and where at least one monitoring value is established by the safety function based on the adjusted measurement value(s) and the measurement value(s) associated with the maximum counter increment.

Abstract: A polyethylene terephthalate aerogel. There is provided a polyethylene terephthalate (PET) aerogel comprising a porous network of cross-linked recycled PET fibers, wherein the PET aerogel has a thermal conductivity of 0.030-0.050 W/m K. There is also provided a method of forming the PET aerogel.

Abstract: Improved soles for shoes, in particular sports shoes, are described. A sole for a shoe, in particular a sports shoe, is provided that includes a midsole with randomly arranged particles of an expanded material. The sole further includes an element having a higher deformation stiffness in at least one direction than the expanded material. The material of the midsole at least partially surrounds the element.

Abstract: Described are methods and systems are for generating dynamic conversational queries. For example, as opposed to being a simply reactive system, the methods and systems herein provide means for actively determining a user's intent and generating a dynamic query based on the determined user intent. Moreover, these methods and systems generate these queries in a conversational environment.

Abstract: Systems and methods are disclosed for gradual and/or gradient-based pruning for language-specific multilingual neural machine translation. For example, a method may include determining a pruning mask for weights of a multilingual machine translation model based on a first pruning threshold, wherein the pruning mask includes at least one entry set to zero and at least one entry set to one; training the multilingual machine translation model, while applying the pruning mask to the multilingual machine translation model, for translation between a language pair based on training examples from a bilingual translation corpus; updating the pruning mask based on a second pruning threshold; and training the multilingual machine translation model, while applying the updated pruning mask to the multilingual machine translation model, for translation between the language pair based on training examples from the bilingual translation corpus. For example, the weights may be pruned based on gradient-based scores.

Abstract: Systems and techniques are provided for coding video data based on an optical flow correction and a residual correction. For example, a decoding device can obtain a frame of encoded video data associated with an input frame, the frame of encoded video data including an optical flow correction and a residual correction. A predicted optical flow can be generated based on one or more reference frames and a reference optical flow. A corrected prediction frame can be generated based on the predicted optical flow and the optical flow correction. A predicted residual can be generated based on at least the corrected prediction frame and a first reference frame included in the one or more reference frames. The decoding device can generate a reconstructed input frame based on the corrected prediction frame, the predicted residual, and the residual correction.

Abstract: Described are methods for the manufacture of a plastic component, in particular a cushioning element for sports apparel, a plastic component manufactured with such a method, for example a sole or a part of a sole for a shoe, and a shoe with such a sole. According to an aspect of the invention, a method for the manufacture of a plastic component, in particular a cushioning element for sports apparel, is provided which includes loading a mold with a first material which includes particles of an expanded material, and, during loading the mold, pre-heating the particles by supplying energy, wherein the energy is supplied in the form of at least one electromagnetic field.

Abstract: Described are methods for manufacturing a plastic component, in particular a cushioning element for sports apparel, a plastic component manufactured with such methods, for example a sole or a part of a sole for a shoe, and a shoe with such a sole. The method for the manufacture of a plastic component includes loading a mold with a first material includes particles of an expanded material and fusing the surfaces of the particles by supplying energy. The energy is supplied in the form of at least one electromagnetic field.

Abstract: A method and system based on machine learning to create self-adapting analog circuits adapted to change their internal components on-the-fly in response to changes in process, voltage, and temperature to re-tune the electrical characteristics back to nominal specified values is disclose. The method and system herein comprise of designing the analog circuit, generating simulation data for machine learning, creating a full query database, creating and training, using simulation results, a machine learning (ML) model of the circuit and applying the ML model to infer the required changes to internal components of the analog circuit in response to changes in P, V, and T conditions. With this method and system, evaluation of the adverse effects of PVT changes, decision on internal circuit changes, and realization of requisite design changes are performed by the computer system solely within a ML data domain, in a time and resource efficient manner.

Abstract: Systems and techniques are described herein for processing video data. For example, a machine-learning based stereo video coding system can obtain video data including at least a right-view image of a right view of a scene and a left-view image of a left view of the scene. The machine-learning based stereo video coding system can compress the right-view image and the left-view image in parallel to generate a latent representation of the right-view image and the left-view image. The right-view image and the left-view image can be compressed in parallel based on inter-view information between the right-view image and the left-view image, determined using one or more parallel autoencoders.

Abstract: Systems and techniques are described herein for processing video data. For example, an encoding device can obtain a sequence of video data and determine a minimum value in the sequence of video data. The encoding device can, based on the minimum value, identify positions in the sequence of video data associated with entry points for individually entropy codable parcels of a parallel entropy codable sequence of video data. The encoding device can generate the parallel entropy codable sequence of video data. The encoding device can further generate an index for the parallel entropy codable sequence of video data, the index identifying the individually entropy codable parcels within the parallel entropy codable sequence of video data.

Abstract: A processor-implemented method for image compression using an artificial neural network (ANN) includes receiving, at an encoder of the ANN, an image and a spatial segmentation map corresponding to the image. The spatial segmentation map indicates one or more regions of interest. The encoder compresses the image according to a controllable spatial bit allocation. The controllable spatial bit allocation is based on a learned quantization bin size.

Abstract: Described are methods and systems are for generating dynamic conversational queries. For example, as opposed to being a simply reactive system, the methods and systems herein provide means for actively determining a user's intent and generating a dynamic query based on the determined user intent. Moreover, these methods and systems generate these queries in a conversational environment.

Abstract: The present invention relates to an enhanced method for the manufacture of a plastic component (135), in particular a cushioning element for sports apparel, the method comprising: opening a mold (100) by a predetermined amount into a loading position, wherein the mold comprises at least two mold parts (110, 112) and wherein the amount by which the mold is opened influences an available loading volume of the mold, loading a material comprising expanded particles (130) into the loading volume, closing the mold into a closed position, wherein during closing of the mold the mold parts are moved together over different distances (140) in different areas of the mold, compressing the expanded particles by closing the mold and fusing at least the surfaces of the expanded particles to mold the plastic component.

Abstract: Methods and systems are described herein for generating dynamic conversational responses. For example, dynamic conversational responses may facilitate an interactive exchange with users. Therefore, the methods and systems used specialized methods to enriched data that may be indicative of a user's intent prior to processing that data through the machine learning model, as well as a specialized architecture for the machine learning models that take advantage of the user interface format.

Abstract: Improved soles for shoes, in particular sports shoes, are described. A sole for a shoe, in particular a sports shoe, is provided that includes a midsole with randomly arranged particles of an expanded material. The sole further includes an element having a higher deformation stiffness in at least one direction than the expanded material. The material of the midsole at least partially surrounds the element.

Abstract: The present invention relates to a method for processing data executed by a computing apparatus including at least a storage and a processor in a human-computer interface system using electroencephalography (EEG) data and eye tracking (ET) data, the method comprising following steps: representing through the control interface, a set of list of keys which is associated with user actions; collecting and processing the two EEG and ET dataset, to determine whether the user is looking at a certain key or not and the concentration state of the user while looking at that key, if yes then activating a time counting parameter t with an increasing rate of the time counting parameter ?>0 and an adjusting rate ?T; confirming whether the user wants to select that key or not, based on a time constraint defined as t?dwell, wherein dwell is a known parameter according to a conventional method corresponding to a required time duration that the user has to look at a key to select that key.

Abstract: Methods and systems are described for generating dynamic interface options using machine learning models. The dynamic interface options may be generated in real time and reflect the likely goals and/or intents of a user. The machine learning model may provide these features by interpreting multi-modal feature inputs. For example, the machine learning model may include a first machine learning model, wherein the first machine learning model comprises a convolutional neural network, and a second machine learning model, wherein the second machine learning model performs a Weight of Evidence (WOE) analysis.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a IT machine learning model. One of the methods includes receiving training data comprising a plurality of unlabeled training inputs and a plurality of labeled training inputs; generating augmented training data, comprising generating, for each of the plurality of unlabeled training inputs, a respective augmented training input by applying a data augmentation technique to the unlabeled training input; and training the machine learning model on the augmented training data. In particular, but not exclusively, the model may be trained for perceptual tasks (e.g. tasks relating to vision or speech).

Abstract: Provided herein are systems and methods for using multi-modal regression to predict customer intent to contact a merchant. Multi-modal data including numerical data and unstructured data are extracted from customer interactions with the merchant. Features of the numerical data and the unstructured data are separately extracted and classified using techniques specific to the data types. The features for each type are then separately used to predict probabilities of customer intent. A neural network is used to combine the predictions into a single set of estimates of customer intent. This set of estimates of customer intents is used to estimate a probability that the customer will contact the merchant. The customer is then contacted based on the estimate.