检索结果-内蒙古大学图书馆

Quantum algorithm for nonlinear differential equations

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Lloyd, Seth de Palma, Giacomo Gokler, Can Kiani, Bobak Liu, Zi-Wen Tennie, Felix Palmer, Tim Marvian, Milad Department of Mechanical Engineering MIT United States Research Lab for Electronics MIT United States Engineering And Applied Sciences Harvard University United States Department of Electrical Engineering and Computer Science MIT United States Perimeter Institute Department of Electrical and Computer Engineering Department of Physics UNM Department of Physics University of Oxford United Kingdom

Quantum computers are known to provide an exponential advantage over classical computers for the solution of linear differential equations in high-dimensional spaces. Here, we present a quantum algorithm for the solution of nonlinear differential equations. The quantum algorithm provides an exponential advantage over classical algorithms for solving nonlinear differential equations. Potential applications include the Navier-Stokes equation, plasma hydrodynamics, epidemiology, and more. © 2020, CC-BY.

关键词： Nonlinear equations

Inverse Design of Perfectly-Matched Metamaterials Via Circuit-Based Surrogate Models and the Adjoint Method

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Thakkar, Shrey Ruiz-García, Jorge Szymanski, Luke Gok, Gurkan Grbic, Anthony Radiation Laboratory Department of Electrical Engineering and Computer Science University of Michigan Ann ArborMI48109 United States MIT Lincoln Laboratory Lexington United States RTX Technology Research Center East HartfordCT06108 United States

In this work, perfectly-matched metamaterials (PMMs) are described and combined with inverse design to realize broadband devices. PMMs are discretized metamaterials with anisotropic unit cells selected from a constrained design space, referred to as perfectly-matched media. PMMs exhibit the unique property that all their unit cells are impedance-matched to each other as well as to the host medium they are embedded within under all excitations. As a result, PMM devices rely on reflectionless refractive effects to achieve a prescribed function. This property enables true time delay performance and promises broadband capabilities. Two design examples are presented to demonstrate the potential of inverse-designed PMMs: a compact, broadband beam-collimator with a prescribed amplitude taper and a multi-input multi-output beamformer exhibiting zero scan loss. Copyright © 2025, The Authors. All rights reserved.

关键词： Metamaterials

Self-focusing virtual screening with active design space pruning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Graff, David E. Aldeghi, Matteo Morrone, Joseph A. Jordan, Kirk E. Pyzer-Knapp, Edward O. Coley, Connor W. Department of Chemistry and Chemical Biology Harvard University CambridgeMA02138 United States Department of Chemical Engineering MIT CambridgeMA02142 United States Computational Biology Center IBM Thomas J. Watson Research Center Yorktown HeightsNY10594 United States IBM Thomas J. Watson Research Center CambridgeMA02142 United States IBM Research Europe Daresbury United Kingdom Department of Electrical Engineering and Computer Science MIT CambridgeMA02142 United States

High-throughput virtual screening is an indispensable technique utilized in the discovery of small molecules. In cases where the library of molecules is exceedingly large, the cost of an exhaustive virtual screen may be prohibitive. Model-guided optimization has been employed to lower these costs through dramatic increases in sample efficiency compared to random selection. However, these techniques introduce new costs to the workflow through the surrogate model training and inference steps. In this study, we propose an extension to the framework of model-guided optimization that mitigates inferences costs using a technique we refer to as design space pruning (DSP), which irreversibly removes poor-performing candidates from consideration. We study the application of DSP to a variety of optimization tasks and observe significant reductions in overhead costs while exhibiting similar performance to the baseline optimization. DSP represents an attractive extension of model-guided optimization that can limit overhead costs in optimization settings where these costs are non-negligible relative to objective costs, such as docking. Copyright © 2022, The Authors. All rights reserved.

关键词： Molecules

Patient contrastive learning: A performant, expressive, and practical approach to ECG modeling

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Diamant, Nathaniel Reinertsen, Erik Song, Steven Aguirre, Aaron Stultz, Collin Batra, Puneet Research Laboratory of Electronics MIT Data Sciences Platform Broad Institute of MIT and Harvard Division of Cardiology Massachusetts General Hospital Center for Systems Biology Massachusetts General Hospital Research Institute Harvard Medical School Wellman Center for Photomedicine Massachusetts General Hospital Research Institute Harvard Medical School Department of Electrical Engineering and Computer Science MIT Harvard-MIT Division of Health Sciences and Technology

Supervised machine learning applications in health care are often limited due to a scarcity of labeled training data. To mitigate this effect of small sample size, we introduce a pre-training approach, Patient Contrastive Learning of Representations (PCLR), which creates latent representations of ECGs from a large number of unlabeled examples. The resulting representations are expressive, performant, and practical across a wide spectrum of clinical tasks. We develop PCLR using a large health care system with over 3.2 million 12-lead ECGs, and demonstrate substantial improvements across multiple new tasks when there are fewer than 5,000 labels. We release our model to extract ECG representations at https://***/broadinstitute/ml4h/tree/master/model_zoo/PCLR. Copyright © 2021, The Authors. All rights reserved.

关键词： Supervised learning

Building Human Values into Recommender Systems: An Interdisciplinary Synthesis

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Stray, Jonathan Halevy, Alon Assar, Parisa Hadfield-Menell, Dylan Boutilier, Craig Ashar, Amar Beattie, Lex Ekstrand, Michael Leibowicz, Claire Sehat, Connie Moon Johansen, Sara Kerlin, Lianne Vickrey, David Singh, Spandana Vrijenhoek, Sanne Zhang, Amy Andrus, McKane Helberger, Natali Proutskova, Polina mitra, Tanushree Vasan, Nina Center for Human-Compatible Artificial Intelligence UC Berkeley United States Meta AI United States Department of Electrical Engineering and Computer Science MIT United States Google Research United States Spotify Inc. Computer Science Department Boise State University United States Partnership on AI Hacks/Hackers Brainstorm The Stanford Lab for Mental Health Innovation Department of Psychiatry and Behavioral Sciences Stanford University United States BBC United Kingdom Meta United States New America's Open Technology Institute United States Faculty of Law Information Law University of Amsterdam Netherlands Allen School of Computer Science & Engineering University of Washington United States Institute for Information Law University of Amsterdam Netherlands Information and Computer Science University of Washington United States

Recommender systems are the algorithms which select, filter, and personalize content across many of the world's largest platforms and apps. As such, their positive and negative effects on individuals and on societies have been extensively theorized and studied. Our overarching question is how to ensure that recommender systems enact the values of the individuals and societies that they serve. Addressing this question in a principled fashion requires technical knowledge of recommender design and operation, and also critically depends on insights from diverse fields including social science, ethics, economics, psychology, policy and law. This paper is a multidisciplinary effort to synthesize theory and practice from different perspectives, with the goal of providing a shared language, articulating current design approaches, and identifying open problems. It is not a comprehensive survey of this large space, but a set of highlights identified by our diverse author cohort. We collect a set of values that seem most relevant to recommender systems operating across different domains, then examine them from the perspectives of current industry practice, measurement, product design, and policy approaches. Important open problems include multi-stakeholder processes for defining values and resolving trade-offs, better values-driven measurements, recommender controls that people use, non-behavioral algorithmic feedback, optimization for long-term outcomes, causal inference of recommender effects, academic-industry research collaborations, and interdisciplinary policy-making. Copyright © 2022, The Authors. All rights reserved.

关键词： Recommender systems

One- and two-dimensional quantum lattice algorithms for Maxwell equations in inhomogeneous scalar dielectric media. II: Simulations

学校读者我要写书评

暂无评论

Radiation Effects and Defects in Solids 2021年第1-2期176卷 64-72页

作者： George Vahala Min Soe Linda Vahala Abhay K. Ram a Department of Physics William & Mary Williamsburg VA USA b Department of Mathematics and Physical Sciences Rogers State University Claremore OK USA c Department of Electrical & Computer Engineering Old Dominion University Norfolk VA USA d Plasma Science and Fusion Center MIT Cambridge MA USA

Long-time quantum lattice algorithm (QLA) simulations are performed for the multiple reflection–transmission of an initial electromagnetic pulse propagating normally to a boundary layer region joining two media of different refractive index. For these one-dimensional (1D) simulations, there is excellent agreement betweenx-,y- andz-representations, as well as very good agreement with nearly all the standard plane wave boundary condition results for reflection and transmission off a dielectric discontinuity. In the QLA simulation, no boundary conditions are imposed at the continuous, but sharply increasing, dielectric boundary layers. Two-dimensional (2D) QLA scattering simulations in thex–zplane are performed for an electromagnetic pulse interacting with a conical dielectric obstacle for the 8–16 qubit model.

关键词： Maxwell equations dielectric media simulations qubits quantum lattice algorithm

Interface engineering Enabled Low Temperature Growth of Magnetic Insulator on Topological Insulator

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Bhattacharjee, Nirjhar Mahalingam, Krishnamurthy Will-Cole, Alexandria Wei, Yuyi Fedorko, Adrian Bowers, Cynthia T. Page, Michael McConney, Michael Heiman, Don Sun, Nian Xiang Northeastern University Department of Electrical and Computer Engineering BostonMA02115 United States Northeastern University Department of Physics BostonMA02115 United States Plasma Science and Fusion Center MIT CambridgeMA02139 United States Air Force Research Laboratory Nano-Electronic Materials Branch Wright Patterson Air Force Base OH05433 United States

Combining topological insulators (TIs) and magnetic materials in heterostructures is crucial for advancing spin-based electronics. Magnetic insulators (MIs) can be deposited on TIs using the spin-spray process, which is a unique non-vacuum, low-temperature growth process. TIs have highly reactive surfaces that oxidize upon exposure to atmosphere, making it challenging to grow spin-spray ferrites on TIs. In this work, it is demonstrated that a thin titanium capping layer on TI, followed by oxidation in atmosphere to produce a thin TiOx interfacial layer, protects the TI surface, without significantly compromising spin transport from the magnetic material across the TiOx to the TI surface states. First, it was demonstrated that in Bi2Te3/TiOx/Ni80Fe20 heterostructures that TiOx provided an excellent barrier against diffusion of magnetic species, yet maintained a large spin-pumping effect. Second, the TiOx was also used as a protective capping layer on Bi2Te3, followed by the spin-spray growth of the MI, NixZnyFe2O4 (NZFO). For the thinnest TiOx barriers, Bi2Te3/TiOx/NZFO samples had AFM disordered interfacial layer because of diffusion. With increasing TiOx barrier thickness, the diffusion was reduced, but still maintained strong interfacial spin-pumping interaction. These experimental results demonstrate a novel method of low-temperature growth of magnetic insulators on TIs enabled by interface engineering. © 2022, CC BY.

关键词： Temperature

Quantifying disparities in intimate partner violence: a machine learning method to correct for underreporting

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Shanmugam, Divya Hou, Kaihua Pierson, Emma Department of Electrical Engineering and Computer Science MIT CambridgeMA02139 United States Malone Center of Engineering in Healthcare Johns Hopkins University School of Medicine BaltimoreMD United States Department of Computer Science Cornell Tech New YorkNY10044 United States Department of Population Health Sciences Weill Cornell Medical College New YorkNY10021 United States

The first step towards reducing the pervasive disparities in women's health is to quantify them. Accurate estimates of the relative prevalence across groups - capturing, for example, that a condition affects Black women more frequently than white women - facilitate effective and equitable health policy which prioritizes groups who are disproportionately affected by a condition. However, it is difficult to estimate relative prevalence when a health condition is underreported, as many women's health conditions are. In this work, we present PURPLE, a method for accurately estimating the relative prevalence of underreported health conditions which builds upon the positive unlabeled machine learning framework. We show that under a commonly made assumption-that the probability of having a health condition given a set of symptoms remains constant across groups-we can recover the relative prevalence, even without restrictive assumptions commonly made in positive unlabeled learning and even if it is impossible to recover the absolute prevalence. We conduct experiments on synthetic and real health data which demonstrate PURPLE's ability to recover the relative prevalence more accurately than do previous methods. We then use PURPLE to quantify the relative prevalence of intimate partner violence (IPV) in two large emergency department datasets. We find higher prevalences of IPV among patients who are on Medicaid, not legally married, in lower-income zipcodes, in metropolitan counties, and non-white. We show that correcting for underreporting is important to accurately quantify these disparities, and that failing to do so yields less plausible estimates. Our method is broadly applicable to underreported conditions in women's health, as well as to gender biases beyond healthcare. © 2021, CC BY.

关键词： Machine learning

Machine Learning-Aided At-Line Detection of Bacterial marker NA for Cell Manufacturing

学校读者我要写书评

暂无评论

Machine Learning-Aided At-Line Detection of Bacterial marker...

Laser science, LS 2021 - Part of Frontiers in Optics, FiO 2021

作者： Huang, Jiayi Ahmed, Thiara Sana Baranski, Maciej Lee, Elizabeth Chelvam, Shruthi Pandi Wu, Ying Ying Ram, Rajeev Rice, Scott Yong, Derrick Springs, Stacy Critical Analytics for Manufacturing Personalised-medicine Singapore-MIT Alliance for Research and Technology Centre 1 Create Way 138602 Singapore School of Chemical and Biomedical Engineering Nanyang Technological University 62 Nanyang Dr Singapore637459 Singapore Bioprocessing Technology Institute Agency for Science Technology and Research 20 Biopolis Way 138668 Singapore Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States Singapore Centre for Environmental Life Sciences Engineering 60 Nanyang Dr 637551 Singapore Singapore Institute of Manufacturing Technology Agency for Science Technology and Research 2 Fusionopolis Way 138634 Singapore Center for Biomedical Innovation Massachusetts Institute of Technology CambridgeMA02139 United States

We show a machine learning-aided UV spectroscopy-based method using aseptic instrumentation to detect metabolite NA, a marker of microbial contamination. This potentially enables rapid, at-line microbial contamination... 详细信息

ISBN: (纸本)9781557528209

关键词： Contamination