检索结果-内蒙古大学图书馆

Nonparametric density estimation with adversarial losses

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Singh, Shashank Uppal, Ananya Li, Boyue Li, Chun-Liang Zaheer, Manzil Póczos, Barnabás Machine Learning Department Department of Statistics and Data Science Department of Mathematical Sciences Language Technologies Institute Carnegie Mellon University

We study minimax convergence rates of nonparametric density estimation under a large class of loss functions called "adversarial losses", which, besides classical Lp losses, includes maximum mean discrepancy (MMD), Wasserstein distance, and total variation distance. These losses are closely related to the losses encoded by discriminator networks in generative adversarial networks (GANs). In a general framework, we study how the choice of loss and the assumed smoothness of the underlying density together determine the minimax rate. We also discuss implications for training GANs based on deep ReLU networks, and more general connections to learning implicit generative models in a minimax statistical sense. Copyright © 2018, The Authors. All rights reserved.

关键词： Generative adversarial networks

BraTS-PEDs: Results of the Multi-Consortium International Pediatric Brain Tumor Segmentation Challenge 2023

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kazerooni, Anahita Fathi Khalili, Nastaran Liu, Xinyang Haldar, Debanjan Jiang, Zhifan Zapaishchykova, Anna Pavaine, Julija Shah, Lubdha M. Jones, Blaise V. Sheth, Nakul Prabhu, Sanjay P. McAllister, Aaron S. Tu, Wenxin Nandolia, Khanak K. Rodriguez, Andres F. Shaikh, Ibraheem Salman Sanchez-Montano, Mariana Lai, Hollie Anne Adewole, Maruf Albrecht, Jake Anazodo, Udunna Anderson, Hannah Anwar, Syed Muhammed Aristizabal, Alejandro Bagheri, Sina Baid, Ujjwal Bergquist, Timothy Borja, Austin J. Calabrese, Evan Chung, Verena Conte, Gian-Marco Eddy, James Ezhov, Ivan Familiar, Ariana M. Farahani, Keyvan Gandhi, Deep Gottipati, Anurag Haldar, Shuvanjan Iglesias, Juan Eugenio Janas, Anastasia Elaine, Elaine Karargyris, Alexandros Kassem, Hasan Khalili, Neda Kofler, Florian LaBella, Dominic Van Leemput, Koen Li, Hongwei B. Maleki, Nazanin Meier, Zeke Menze, Bjoern Moawad, Ahmed W. Pati, Sarthak Piraud, Marie Poussaint, Tina Reitman, Zachary J. Rudie, Jeffrey D. Saluja, Rachit Sheller, MIcah Shinohara, Russell Takeshi Viswanathan, Karthik Wang, Chunhao Wiestler, Benedikt Wiggins, Walter F. Davatzikos, Christos Storm, Phillip B. Bornhorst, Miriam Packer, Roger Hummel, Trent de Blank, Peter Hoffman, Lindsey Aboian, Mariam Nabavizadeh, Ali Ware, Jeffrey B. Kann, Benjamin H. Rood, Brian Resnick, Adam Bakas, Spyridon Vossough, Arastoo Linguraru, Marius George The Children’s Hospital of Philadelphia PA United States Sheikh Zayed Institute for Pediatric Surgical Innovation Children’s National Hospital WashingtonDC United States Department of Neurosurgery Thomas Jefferson University Hospital PA United States Dana-Farber Brigham Cancer Center Boston Children’s Hospital BostonMA United States Department of Paediatric Radiology Royal Manchester Children’s Hospital Manchester University Hospitals NHS Foundation Trust University of Manchester Manchester United Kingdom University of Utah UT United States Cincinnati Children’s Hospital Medical Center OH United States Department of Radiology The Children’s Hospital of Philadelphia PA United States Division of Neuroradiology Department of Radiology Boston Children’s Hospital Harvard Medical School BostonMA United States Department of Radiology Nationwide Children’s Hospital OH United States College of Arts and Sciences University of Pennsylvania PA United States Department of Diagnostic and Interventional Radiology All India Institute of Medical Sciences Rishikesh India Department of Diagnostic & Interventional Imaging Neuroradiology Section The University of Texas Health HoustonTX United States Beth Israel Deaconess Medical Center Harvard Medical School MA United States Unidad de Patologia Clinica:Guadalajara Jalisco Mexico Department of Radiology Children’s Health Orange County CA United States Lab Crestview Radiology Lagos Nigeria Sage Bionetworks United States McGill University MontrealQC Canada Department of Radiology Perelman School of Medicine University of Pennsylvania PhiladelphiaPA United States MLCommons United States Division of Computational Pathology Department of Pathology and Laboratory Medicine Indiana University School of Medicine IndianapolisIN United States Department of Neurosurgery The University of Southern California CA United States Department of Radiology Duke University Medical Center DurhamNC United States Mayo Clinic MN

Pediatric central nervous system tumors are the leading cause of cancer-related deaths in children. The five-year survival rate for high-grade glioma in children is less than 20%. The development of new treatments is dependent upon multi-institutional collaborative clinical trials requiring reproducible and accurate centralized response assessment. We present the results of the BraTS-PEDs 2023 challenge, the first Brain Tumor Segmentation (BraTS) challenge focused on pediatric brain tumors. This challenge utilized data acquired from multiple international consortia dedicated to pediatric neuro-oncology and clinical trials. BraTS-PEDs 2023 aimed to evaluate volumetric segmentation algorithms for pediatric brain gliomas from magnetic resonance imaging using standardized quantitative performance evaluation metrics employed across the BraTS 2023 challenges. The top-performing AI approaches for pediatric tumor analysis included ensembles of nnU-Net and Swin UNETR, Auto3DSeg, or nnU-Net with a self-supervised framework. The BraTS-PEDs 2023 challenge fostered collaboration between clinicians (neuro-oncologists, neuroradiologists) and AI/imaging scientists, promoting faster data sharing and the development of automated volumetric analysis techniques. These advancements could significantly benefit clinical trials and improve the care of children with brain tumors. © 2024, CC BY.

关键词： Pediatrics

Roaming leads to amino acid photodamage: A deep learning study of tyrosine

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Westermayr, Julia Gastegger, Michael Vörös, Dora Panzenboeck, Lisa Joerg, Florian González, Leticia Marquetand, Philipp University of Warwick Department of Chemistry Gibbet Hill Rd CoventryCV4 7AL United Kingdom Technical University of Berlin Machine Learning Group Berlin10587 Germany University of Vienna Faculty of Chemistry Institute of Theoretical Chemistry Währinger Str. 17 Vienna1090 Austria University of Vienna Vienna Research Platform on Accelerating Photoreaction Discovery Währinger Str. 17 Vienna1090 Austria University of Vienna Faculty of Chemistry Data Science @ Uni Vienna Währinger Str. 29 Vienna1090 Austria

Although the amino acid tyrosine is among the main building blocks of life, its photochemistry is not fully understood. Traditional theoretical simulations are neither accurate enough, nor computationally efficient to provide the missing puzzle pieces to the experimentally observed signatures obtained via time-resolved pump-probe spectroscopy or mass spectroscopy. In this work, we go beyond the realms of possibility with conventional quantum chemical methods and develop as well as apply a new technique to shed light on the photochemistry of tyrosine. By doing so, we discover roaming atoms in tyrosine, which is the first time such a reaction is discovered in biology. Our findings suggest that roaming atoms are radicals that could play a fundamental role in the photochemistry of peptides and proteins, offering a new perspective. Our novel method is based on deep learning, leverages the physics underlying the data, and combines different levels of theory. This combination of methods to obtain an accurate picture of excited molecules could shape how we study photochemical systems in the future and how we can overcome the current limitations that we face when applying quantum chemical methods. Copyright © 2021, The Authors. All rights reserved.

关键词： Amino acids

Depth selection for deep ReLU nets in feature extraction and generalization

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Han, Zhi Yu, Siquan Lin, Shao-Bo Zhou, Ding-Xuan State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang China School of Information Science and Engineering Northeastern University Shenyang China Center of Intelligent Decision-Making and Machine Learning School of Management Xi'an Jiaotong University Xi'an China School of Data Science Department of Mathematics City University of Hong Kong Hong Kong

Deep learning is recognized to be capable of discovering deep features for representation learning and pattern recognition without requiring elegant feature engineering techniques by taking advantage of human ingenuity and prior knowledge. Thus it has triggered enormous research activities in machine learning and pattern recognition. One of the most important challenge of deep learning is to figure out relations between a feature and the depth of deep neural networks (deep nets for short) to reflect the necessity of depth. Our purpose is to quantify this feature-depth correspondence in feature extraction and generalization. We present the adaptivity of features to depths and vice-verse via showing a depth-parameter trade-off in extracting both single feature and composite features. Based on these results, we prove that implementing the classical empirical risk minimization on deep nets can achieve the optimal generalization performance for numerous learning tasks. Our theoretical results are verified by a series of numerical experiments including toy simulations and a real application of earthquake seismic intensity prediction. Copyright © 2020, The Authors. All rights reserved.

关键词： Feature extraction

The Brain Tumor Segmentation in Pediatrics (BraTS-PEDs) Challenge: Focus on Pediatrics (CBTN-CONNECT-DIPGR-ASNR-MICCAI BraTS-PEDs)

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kazerooni, Anahita Fathi Khalili, Nastaran Liu, Xinyang Gandhi, Deep Jiang, Zhifan Anwar, Syed Muhammed Albrecht, Jake Adewole, Maruf Anazodo, Udunna Anderson, Hannah Baid, Ujjwal Bergquist, Timothy Borja, Austin J. Calabrese, Evan Chung, Verena Conte, Gian-Marco Dako, Farouk Eddy, James Ezhov, Ivan Familiar, Ariana Farahani, Keyvan Franson, Andrea Gottipati, Anurag Haldar, Shuvanjan Iglesias, Juan Eugenio Janas, Anastasia Johansen, Elaine Jones, Blaise V. Khalili, Neda Kofler, Florian LaBella, Dominic Lai, Hollie Anne Van Leemput, Koen Li, Hongwei Bran Maleki, Nazanin McAllister, Aaron S. Meier, Zeke Menze, Bjoern Moawad, Ahmed W. Nandolia, Khanak K. Pavaine, Julija Piraud, Marie Poussaint, Tina Prabhu, Sanjay P. Reitman, Zachary Rudie, Jeffrey D. Sanchez-Montano, Mariana Shaikh, Ibraheem Salman Sheth, Nakul Tu, Wenxin Wang, Chunhao Ware, Jeffrey B. Wiestler, Benedikt Zapaishchykova, Anna Bornhorst, Miriam Deutsch, Michelle Fouladi, Maryam Lazow, Margot Mikael, Leonie Hummel, Trent Kann, Benjamin de Blank, Peter Hoffman, Lindsey Aboian, Mariam Nabavizadeh, Ali Packer, Roger Bakas, Spyridon Resnick, Adam Rood, Brian Vossough, Arastoo Linguraru, Marius George Children’s Hospital of Philadelphia PhiladelphiaPA United States Department of Neurosurgery University of Pennsylvania PhiladelphiaPA United States University of Pennsylvania PhiladelphiaPA United States Sheikh Zayed Institute for Pediatric Surgical Innovation Children’s National Hospital WashingtonDC United States Sage Bionetworks United States Lab Crestview Radiology Lagos Nigeria McGill University MontrealQC Canada Department of Radiology Perelman School of Medicine University of Pennsylvania PhiladelphiaPA United States Division of Computational Pathology Department of Pathology and Laboratory Medicine Indiana University School of Medicine IndianapolisIN United States Department of Neurosurgery The University of Southern California CA United States Department of Radiology Duke University Medical Center DurhamNC United States Mayo Clinic MN United States Center for Global Health Perelman School of Medicine University of Pennsylvania PhiladelphiaPA United States Department of Informatics Technical University Munich Germany TranslaTUM - Central Institute for Translational Cancer Research Technical University of Munich Germany Cancer Imaging Program National Cancer Institute National Institutes of Health BethesdaMD United States C. S. Mott Children’s Hospital University of Michigan MI United States Biomedical Engineering Rutgers University New BrunswickNJ United States Athinoula A Martinos Center for Biomedical Imaging Massachusetts General Hospital BostonMA United States Yale University New HavenCT United States PrecisionFDA U.S. Food and Drug Administration Silver SpringMD United States Cincinnati Children’s Hospital Medical Center OH United States Helmholtz AI Helmholtz Munich Germany Department of Radiation Oncology Duke University Medical Center DurhamNC United States Department of Radiology Children’s Health Orange County CA United States Department of Applied Mathematics and Computer Science Technical University of De

Pediatric tumors of the central nervous system are the most common cause of cancer-related death in children. The five-year survival rate for high-grade gliomas in children is less than 20%. Due to their rarity, the diagnosis of these entities is often delayed, their treatment is mainly based on historic treatment concepts, and clinical trials require multi-institutional collaborations. Here we present the CBTN-CONNECT-DIPGR-ASNR-MICCAI BraTS-PEDs challenge, focused on pediatric brain tumors with data acquired across multiple international consortia dedicated to pediatric neuro-oncology and clinical trials. The CBTN-CONNECT-DIPGR-ASNR-MICCAI BraTS-PEDs challenge brings together clinicians and AI/imaging scientists to lead to faster development of automated segmentation techniques that could benefit clinical trials, and ultimately the care of children with brain tumors. © 2024, CC BY.

关键词： Deep learning

Prime-residue-class of uniform charges on the integers

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Spece, Michael Kadane, Joseph B. Department of Statistics & Data Science and Machine Learning Department Carnegie Mellon University PittsburghPA15213 United States Department of Statistics & Data Science Carnegie Mellon University PittsburghPA15213 United States

There is a probability charge on the power set of the integers that gives probability 1/p to every residue class modulo a prime p. There exists such a charge that gives probability w to the set of prime numbers iff w ∈ [0,1/2]. Similarly, there is such a charge that gives probability x to a residue class modulo c, where c is composite, iff x ∈ [0,1/y], where y is the largest prime factor of c. Copyright © 2018, The Authors. All rights reserved.

关键词： C (programming language)

Optimal Neural Summarisation for Full-Field Weak Lensing Cosmological Implicit Inference

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lanzieri, Denise Zeghal, Justine Makinen, T. Lucas Boucaud, Alexandre Starck, Jean-Luc Lanusse, François Université Paris Cité Université Paris-Saclay CEA CNRS AIM Gif-sur-YvetteF-91191 France Université Paris Cité CNRS Astroparticule et Cosmologie ParisF-75013 France Astrophysics Group Imperial College London Blackett Laboratory Prince Consort Road LondonSW7 2AZ United Kingdom Université Paris-Saclay Université Paris Cité CEA CNRS AIM Gif-sur-Yvette91191 France Sony Computer Science Laboratories - Rome Joint Initiative CREF-SONY Centro Ricerche Enrico Fermi Via Panisperna 89/A Rome00184 Italy Greece Center for Computational Astrophysics Flatiron Institute 162 5th Ave New YorkNY10010 United States Department of Physics Université de Montréal MontréalH2V 0B3 Canada Mila – Quebec Artificial Intelligence Institute MontréalH2S 3H1 Canada Ciela – Montreal Institute for Astrophysical Data Analysis and Machine Learning MontréalH2V 0B3 Canada

Context. Traditionally, weak lensing cosmological surveys have been analyzed using summary statistics that were either motivated by their analytically tractable likelihoods (e.g. power spectrum), or by their ability to access some higher-order information (e.g. peak counts) but in that case at the cost of requiring a Simulation-Based Inference approach. In both cases, even if these statistics can be very informative, they are not designed nor guaranteed to be statistically sufficient (i.e. to capture all of the cosmological information content of the data). With the rise of deep learning, however, it becomes possible to create summary statistics that are specifically optimized to extract the full cosmological information content of the data. However, a fairly wide range of loss functions have been used in practice in the weak lensing literature to train such neural networks, leading to the natural question of whether a given loss should be preferred and whether sufficient statistics can be achieved in theory and in practice under these different choices. Aims. We compare different neural summarization strategies that have been proposed in the literature, to answer the question of what loss function leads to theoretically optimal summary statistics for performing full-field cosmological inference. In doing so, we aim to provide guidelines and insights to the community to help guide future neural network-based cosmological inference analyses. Methods. We design an experimental setup which allows us to isolate the specific impact of the loss function used to train neural summary statistics on weak lensing data, at fixed neural architecture and Simulation-Based Inference pipeline. To achieve this, we have developed the sbi_lens JAX package, which implements an automatically differentiable lognormal weak lensing simulator and the tools needed to perform explicit full-field inference with a Hamiltonian-Monte Carlo (HMC) sampler over this model. Using sbi_lens, we simulate

关键词： Cosmology

machine learning Force Fields

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Unke, Oliver T. Chmiela, Stefan Sauceda, Huziel E. Gastegger, Michael Poltavsky, Igor Schütt, Kristof T. Tkatchenko, Alexandre Müller, Klaus-Robert Machine Learning Group Technische Universität Berlin Berlin10587 Germany Technische Universität Berlin Berlin10623 Germany BASLEARN BASF-TU joint Lab Technische Universität Berlin Berlin10587 Germany Department of Physics and Materials Science University of Luxembourg Luxembourg CityL-1511 Luxembourg BIFOLD Berlin Institute for the Foundations of Learning and Data Berlin Germany Department of Artificial Intelligence Korea University Anam-dong Seongbuk-gu Seoul02841 Korea Republic of Max Planck Institute for Informatics Stuhlsatzenhausweg Saarbrücken66123 Germany

In recent years, the use of machine learning (ML) in computational chemistry has enabled numerous advances previously out of reach due to the computational complexity of traditional electronic-structure methods. One of the most promising applications is the construction of ML-based force fields (FFs), with the aim to narrow the gap between the accuracy of ab initio methods and the efficiency of classical FFs. The key idea is to learn the statistical relation between chemical structure and potential energy without relying on a preconceived notion of fixed chemical bonds or knowledge about the relevant interactions. Such universal ML approximations are in principle only limited by the quality and quantity of the reference data used to train them. This review gives an overview of applications of ML-FFs and the chemical insights that can be obtained from them. The core concepts underlying ML-FFs are described in detail and a step-by-step guide for constructing and testing them from scratch is given. The text concludes with a discussion of the challenges that remain to be overcome by the next generation of ML-FFs. Copyright © 2020, The Authors. All rights reserved.

关键词： machine learning

A Review of Artificial Fish Swarm Algorithms: Recent Advances and Applications

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Pourpanah, Farhad Wang, Ran Lim, Chee Peng Wang, Xi-Zhao Yazdani, Danial College of Mathematics and Statistics Guangdong Key Lab. of Intelligent Information Processing Shenzhen University China Department of Electrical and Computer Engineering University of Windsor Canada College of Mathematics and Statistics Shenzhen Key Lab. of Advanced Machine Learning and Applications Guangdong Key Lab. of Intelligent Information Processing Shenzhen University China Institute for Intelligent Systems Research and Innovation Deakin University Australia College of Computer Science and Software Engineering Guangdong Key Lab. of Intelligent Information Processing Shenzhen University China School of Computer Science and Engineering Southern University of Science and Technology China

The Artificial Fish Swarm Algorithm (AFSA) is inspired by the ecological behaviors of fish schooling in nature, viz., the preying, swarming and following behaviors. Owing to a number of salient properties, which include flexibility, fast convergence, and insensitivity to the initial parameter settings, the family of AFSA has emerged as an effective Swarm Intelligence (SI) methodology that has been widely applied to solve real-world optimization problems. Since its introduction in 2002, many improved and hybrid AFSA models have been developed to tackle continuous, binary, and combinatorial optimization problems. This paper aims to present a concise review of the continuous AFSA, encompassing the original ASFA, its improvements and hybrid models, as well as their associated applications. We focus on articles published in high-quality journals since 2013. Our review provides insights into AFSA parameters modifications, procedure and sub-functions. The main reasons for these enhancements and the comparison results with other hybrid methods are discussed. In addition, hybrid, multi-objective and dynamic AFSA models that have been proposed to solve continuous optimization problems are elucidated. We also analyse possible AFSA enhancements and highlight future research directions for advancing AFSA-based models. Copyright © 2020, The Authors. All rights reserved.

关键词： Multiobjective optimization