Abstract: Methods, systems, and apparatus for a bi-directional quantum annealing approach to Markov random field networks for machine learning in image analysis. In one aspect, a method includes obtaining training data comprising features extracted from a first set of images; training a deep quantum restricted Boltzmann machine (QRBM) comprising multiple layers using the training data, the training comprising layer-wise training of the multiple layers, wherein training each layer of the multiple layers comprises evaluating a restricted Boltzmann machine (RBM) probability distribution using bi-directional quantum annealing; and validating the trained deep QRBM using test data comprising features extracted from a second set of images.

Abstract: Methods, systems, and apparatus for a bi-directional quantum annealing approach to Markov random field networks for machine learning in image analysis. In one aspect, a method includes obtaining training data comprising features extracted from a first set of images; training a deep quantum restricted Boltzmann machine (QRBM) comprising multiple layers using the training data, the training comprising layer-wise training of the multiple layers, wherein training each layer of the multiple layers comprises evaluating a restricted Boltzmann machine (RBM) probability distribution using bi-directional quantum annealing; and validating the trained deep QRBM using test data comprising features extracted from a second set of images.

Abstract: A device may obtain incomplete geospatial coordinate data associated with a telecommunications network. The device may generate a network graph representation of the telecommunications network based on the incomplete geospatial coordinate data. The device may determine, for candidate sites, a candidate site score based on a quantity of candidate segments to connect the candidate site to a candidate hub. The device may determine, for candidate segments, a candidate segment score based on a quantity of candidate sites that connect to a candidate hub via the candidate segment. The device may determine a prioritization of the candidate segments based on the candidate site scores and the candidate segment scores. The device may generate a recommendation for selecting or ordering the candidate segments. The device may provide the recommendation for display via a user interface.