As convolutional neural networks (CNNs) enable state-of-the-art computer vision applications, their high energy consumption has emerged as a key impediment to their deployment on embedded and mobile devices. Towards efficient image classification under hardware constraints, prior work has proposed adaptive CNNs, i.e., systems of networks with different accuracy and computation characteristics, where a selection scheme adaptively selects the network to be evaluated for each input image. While previous efforts have investigated different network selection schemes, we find that they do not necessarily result in energy savings when deployed on mobile systems. The key limitation of existing methods is that they learn only how data should be processed among the CNNs and not the network architectures, with each network being treated as a blackbox.To address this limitation, we pursue a more powerful design paradigm where the architecture settings of the CNNs are treated as hyper-parameters to be globally optimized. We cast the design of adaptive CNNs as a hyper-parameter optimization problem with respect to energy, accuracy, and communication constraints imposed by the mobile device. To efficiently solve this problem, we adapt Bayesian optimization to the properties of the design space, reaching near-optimal configurations in few tens of function evaluations. Our method reduces the energy consumed for image classification on a mobile device by up to 6×, compared to the best previously published work that uses CNNs as blackboxes. Finally, we evaluate two image classification practices, i.e., classifying all images locally versus over the cloud under energy and communication constraints. * Mitchell Bognar was an intern at Carnegie Mellon University; he is currently a student at Duke University.
Many different digital representations of a building are produced over the course of its lifecycle. These representations contain the metadata required to support different stages of the building, from initial planning and design, to construction and commissioning, through operations, audits, retrofits and maintenance. However, because of differences in the semantics, structure and syntax of these representations, the metadata they contain is not interoperable. We present a novel method for leveraging these representations to create a unified, authoritative Brick metadata model for a building that can be continually maintained over the course of the building lifecycle. A simple synchronization protocol relays inferred Brick metadata from existing metadata sources such as gbXML, Build-ingSync, Project Haystack and Modelica to a central integration server, which merges the metadata into a valid Brick model.
Robust and accurate building energy load forecasting is important for helping building managers and utilities to plan, budget, and strategize energy resources in advance. With recent prevalent adoption of smart-meters in buildings, a significant amount of building energy consumption data became available. Many studies have developed physics-based white box models and data-driven black box models to predict building energy consumption; however, they require extensive prior knowledge about building system, need a large set of training data, or lack robustness to different forecasting scenarios. In this paper, we introduce a new building energy forecasting method based on Gaussian Process Regression (GPR) that incorporates physical insights about load data characteristics to improve accuracy while reducing training requirements. The GPR is a non-parametric regression method that models the data as a joint Gaussian distribution with mean and covariance functions and forecast using the Bayesian updating. We model the covariance function of the GPR to reflect the data patterns in different forecasting horizon scenarios, as prior knowledge. Our method takes advantage of the modeling flexibility and computational efficiency of the GPR while benefiting from the physical insights to further improve the training efficiency and accuracy. We evaluate our method with three field datasets from two university campuses (Carnegie Mellon University and Stanford University) for both short-and long-term load forecasting. The results show that our method performs more accurately, especially when the training dataset is small, compared to other state-of-the-art forecasting models (up to 2.95 times smaller prediction error).
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.