The story goes that Philip II of Spain ran a competition for the design of a monastery in Madrid in the 16th century and hired Italian architect Giacomo Barozzi da Vignola to organize and facilitate the competition. In total, 22 architects responded to the challenge and submitted designs. However, Vignola did not rank the entries and select a winning scheme as planned. Instead, he composed a new design by collaging bits and pieces from competition entries he deemed strong. Vignola then presented his composition to the King, who was impressed and gave him the commission. 1 Of course, such a procedure would certainly break all kinds of ethical norms of today's profession. What if, however, the most ideal solution lies in a collective design approach, that is, in composing high performing parts of separate design entries into a new whole? In this article, we present research on a deep neural network (DNN) or deep learning application that extracts design into essential building blocks-based on functional performance criteria-and recombines them into new designs. Over the last 5 years, research in machine learning has exploded thanks to fast developments in deep learning. DNNs used on a wide range of practical applications, from voice recognition systems, such as Siri
In recent years, new neural network architectures designed to operate on graphstructured data have pushed the state-of-the-art in the field. A large set of these architectures utilize a form of classical random walks to diffuse information. We propose quantum walk neural networks (QWNN), a novel graph neural network architecture based on quantum random walks, the quantum parallel to classical random walks. A QWNN learns a quantum walk on a graph to construct a diffusion operator which can then be applied to graph-structured data. We demonstrate the use of QWNNs on a variety of prediction tasks on graphs involving temperature, biological, and molecular datasets.Preprint. Work in progress. arXiv:1801.05417v2 [quant-ph]
Recent neural networks designed to operate on graph-structured data have proven effective in many domains. These graph neural networks often diffuse information using the spatial structure of the graph. We propose a quantum walk neural network that learns a diffusion operation that is not only dependent on the geometry of the graph but also on the features of the nodes and the learning task. A quantum walk neural network is based on learning the coin operators that determine the behavior of quantum random walks, the quantum parallel to classical random walks. We demonstrate the effectiveness of our method on multiple classification and regression tasks at both node and graph levels.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.