Abstract: Techniques are described herein for training and evaluating machine learning (ML) models for document processing computing applications using generalized vocabulary tokens. In some embodiments, an ML system determines a set of tokens for non-textual content in a plurality of documents. The ML system generates a fixed-length vocabulary that includes the set of tokens for the non-textual content. The ML system further generates for each respective document in a training dataset of documents, a respective feature vector based at least in part on which tokens in the fixed-length vocabulary occur in the respective document. The ML system trains a ML model based at least in part on the respective feature vector for each respective document in the training dataset.

Abstract: Techniques for summarizing lists for machine learning operations are disclosed. In some embodiments, a machine learning system generates feature vectors for a set of items based on varying values among a set of feature attributes. The system further generates, based on the feature vectors a set of clusters and generates a summary vector for a list of items as a function of the distribution of the items within the set of clusters, where the summary vector has a length equal to how many clusters are in the set of clusters. Summary vectors may be generated for a plurality of examples within a training dataset. The system may use the summary vectors to train a machine learning model to estimate unknown labels for new examples.

Abstract: The present disclosure generally relates to techniques for determining a segment of a content distribution plan. More specifically, the present disclosure discloses techniques for determining one or more key-value pairs of a content distribution plan by leveraging a trained machine learning model. A plurality of electronic communications may be generated based on completed key-value pairs with a content distribution plan. The plurality of electronic communications may then be distributed to a plurality of devices within a networked environment.

Abstract: Aspects of the subject disclosure may include, for example, applying a topic detection process to documents to obtain automatically detected topics and groups of automatically detected words, comparing the automatically detected topics with manually determined topics to determine actual purity metrics, determining an error metric based on a measure of deviation between ideal purity metrics and the actual purity metrics, and adjusting a parameter of the topic detection process according to the error metric resulting in an adjusted topic detection process. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, applying a topic detection process to documents to obtain automatically detected topics and groups of automatically detected words, comparing the automatically detected topics with manually determined topics to determine actual purity metrics, determining an error metric based on a measure of deviation between ideal purity metrics and the actual purity metrics, and adjusting a parameter of the topic detection process according to the error metric resulting in an adjusted topic detection process. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, applying a topic detection process to documents to obtain automatically detected topics and groups of automatically detected words, comparing the automatically detected topics with manually determined topics to determine actual purity metrics, determining an error metric based on a measure of deviation between ideal purity metrics and the actual purity metrics, and adjusting a parameter of the topic detection process according to the error metric resulting in an adjusted topic detection process. Other embodiments are disclosed.

Abstract: A method may include a processor of a telecommunication service provider network receiving a training data set including service profile trajectories for subscribers of the network, each including a network service profile of a subscriber over a plurality of time periods, where for a given time period, each network service profile includes indications of whether a subscriber is subscribed to a plurality of network services. The processor may further create a predictive model based upon the training data set to predict whether a subject subscriber will be subscribed to a given network service at a designated future time period, receive a service profile trajectory for the subject subscriber, apply the service profile trajectory to the predictive model to generate a prediction of whether the subject subscriber will be subscribed to the given network service at the designated future time period, and allocate a network resource based upon the prediction.

Abstract: Embodiments of an OFDM receiver and methods for decoding OFDM symbols of two or more data streams with reduced multiplication operations are generally described herein. Other embodiments may be described and claimed. In some embodiments, one or more terms of a modified L2-squared-norm cost function are precomputed and stored for predetermined complex symbol values of one or more tones of OFDM symbols prior to performing a searching process. During the searching process, the cost function is computed using the precomputed terms and received data symbols using shifting and adding operations, rather than multiplication operations. In other embodiments, non-L2-squared-norm cost functions are used.

Abstract: Embodiments of a multicarrier receiver and method for generating soft bits in a multiple-input multiple-output system are generally described herein. In some embodiments, operational parameters for an equalizer and a soft-bit demapper in a multicarrier receiver are determined. Other embodiments may be described and claimed.

Abstract: The effects of interference are mitigated in a wireless system by estimating spatial characteristics of an interfering signal, and using those characteristics in the formation of a spatial equalizer.

Abstract: A line card including: a co-channel estimator and a code selector. The line card is configured to couple to digital subscriber lines to support multi-tone modulation of communications channels thereon. The co-channel estimator is configured to estimate co-channel crosstalk coupling coefficients among selected pairs of the subscriber lines at levels for which the total crosstalk into a selected victim line among the plurality of digital subscriber lines substantially corresponds to the sum of the products of the corresponding crosstalk coupling coefficient for each remaining disturber one of the plurality of subscriber lines and a corresponding substantially unique vector transmitted thereon. The code selector couples to the co-channel estimator. The code selector is configured to select a cross-talk estimation code type and to generate substantially unique code vectors derived there from for injection into selected ones of the of subscriber lines.

Abstract: Embodiments of an OFDM receiver and methods for decoding OFDM symbols of two or more data streams with reduced multiplication operations are generally described herein. Other embodiments may be described and claimed. In some embodiments, one or more terms of a modified L2-squared-norm cost function are precomputed and stored for predetermined complex symbol values of one or more tones of OFDM symbols prior to performing a searching process. During the searching process, the cost function is computed using the precomputed terms and received data symbols using shifting and adding operations, rather than multiplication operations. In other embodiments, non-L2-squared-norm cost functions are used.

Abstract: In one embodiment, an apparatus includes a correlator unit that generates a set of correlator outputs and a codeword selector coupled to the set of correlator outputs to determine a received codeword therefrom. The correlator unit may have reference signals pre-compensated with intra-codeword interference.

Abstract: Embodiments of a multicarrier receiver and method for generating soft bits in a multiple-input multiple-output system are generally described herein. In some embodiments, operational parameters for an equalizer and a soft-bit demapper in a multicarrier receiver are determined. Other embodiments may be described and claimed.

Abstract: The effects of interference are mitigated in a wireless system by estimating spatial characteristics of an interfering signal, and using those characteristics in the formation of a spatial equalizer.

Abstract: In one embodiment, an apparatus includes a correlator unit that generates a set of correlator outputs and a codeword selector coupled to the set of correlator outputs to determine a received codeword therefrom. The correlator unit may have reference signals pre-compensated with intra-codeword interference.