检索结果-内蒙古大学图书馆

Quantum machine learning: a classical perspective

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Ciliberto, Carlo Herbster, Mark Ialongo, Alessandro Davide Pontil, Massimiliano Rocchetto, Andrea Severini, Simone Wossnig, Leonard Department of Computer Science University College London Department of Engineering University of Cambridge Max Planck Institute for Intelligent Systems Tübingen Germany Computational Statistics and Machine Learning Istituto Italiano di Tecnologia Department of Materials University of Oxford Institute of Natural Sciences Shanghai Jiao Tong University Theoretische Physik ETH Zürich

Recently, increased computational power and data availability, as well as algorithmic advances, have led machine learning techniques to impressive results in regression, classification, data-generation and reinforcement learning tasks. Despite these successes, the proximity to the physical limits of chip fabrication alongside the increasing size of datasets are motivating a growing number of researchers to explore the possibility of harnessing the power of quantum computation to speed-up classical machine learning algorithms. Here we review the literature in quantum machine learning and discuss perspectives for a mixed readership of classical machine learning and quantum computation experts. Particular emphasis will be placed on clarifying the limitations of quantum algorithms, how they compare with their best classical counterparts and why quantum resources are expected to provide advantages for learning problems. learning in the presence of noise and certain computationally hard problems in machine learning are identified as promising directions for the field. Practical questions, like how to upload classical data into quantum form, will also be addressed. Copyright © 2017, The Authors. All rights reserved.

关键词： Reinforcement learning

Cosmology with Galaxy Photometry Alone

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hahn, ChangHoon Villaescusa-Navarro, Francisco Melchior, Peter Teyssier, Romain Department of Astrophysical Sciences Princeton University Peyton Hall PrincetonNJ08544 United States Center for Computational Astrophysics Flatiron Institute 162 5th Avenue New YorkNY10010 United States Center for Statistics & Machine Learning Princeton University PrincetonNJ08544 United States Program in Applied and Computational Mathematics Princeton University Fine Hall Washington Road PrincetonNJ08544-1000 United States

We present the first cosmological constraints using only the observed photometry of galaxies. Villaescusa-Navarro et al. (2022b) recently demonstrated that the internal physical properties of a single simulated galaxy contain a significant amount of cosmological information. These physical properties, however, cannot be directly measured from observations. In this work, we present how we can go beyond theoretical demonstrations to infer cosmological constraints from actual galaxy observables (e.g. optical photometry) using neural density estimation and the CAMELS suite of hydrodynamical simulations. We find that the cosmological information in the photometry of a single galaxy is limited. However, we combine the constraining power of photometry from many galaxies using hierarchical population inference and place significant cosmological constraints. With the observed photometry of ∼20,000 NASA-Sloan Atlas galaxies, we constrain (formula presented) and (formula presented). © 2023, CC BY.

关键词： Galaxies

Communication-Avoiding Optimization Methods for Distributed Massive-Scale Sparse Inverse Covariance Estimation

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Koanantakool, Penporn Ali, Alnur Azad, Ariful Buluç, Aydın Morozov, Dmitriy Oliker, Leonid Yelick, Katherine Oh, Sang-Yun Department of Electrical Engineering and Computer Sciences UC Berkeley United States Machine Learning Department Carnegie Mellon University United States Computational Research Division Lawrence Berkeley National Laboratory United States Berkeley Institute for Data Science UC Berkeley United States Department of Statistics and Applied Probability UC Santa Barbara United States Google Brain United States

Across a variety of scientific disciplines, sparse inverse covariance estimation is a popular tool for capturing the underlying dependency relationships in multivariate data. Unfortunately, most estimators are not scalable enough to handle the sizes of modern high-dimensional data sets (often on the order of terabytes), and assume Gaussian samples. To address these deficiencies, we introduce HP-CONCORD, a highly scalable optimization method for estimating a sparse inverse covariance matrix based on a regularized pseudolikelihood framework, without assuming Gaussianity. Our parallel proximal gradient method uses a novel communication-avoiding linear algebra algorithm and runs across a multi-node cluster with up to 1k nodes (24k cores), achieving parallel scalability on problems with up to ≈819 billion parameters (1.28 million dimensions);even on a single node, HP-CONCORD demonstrates scalability, outperforming a state-of-the-art method. We also use HP-CONCORD to estimate the underlying dependency structure of the brain from fMRI data, and use the result to identify functional regions automatically. The results show good agreement with a clustering from the neuroscience literature. Copyright © 2017, The Authors. All rights reserved.

关键词： Scalability

Variational Policy Gradient Method for Reinforcement learning with General Utilities

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhang, Junyu Koppel, Alec Bedi, Amrit Singh Szepesvari, Csaba Wang, Mengdi Department of Industrial and Systems Engineering University of Minnesota MinneapolisMN55455 United States Computational and Information Sciences Directorate US Army Research Laboratory AdelphiMD20783 United States Department of Computer Science DeepMind University of Alberta PrincetonNJ08544 United States Department of Electrical Engineering Center for Statistics and Machine Learning Princeton University Deepmind PrincetonNJ08544 United States

In recent years, reinforcement learning (RL) systems with general goals beyond a cumulative sum of rewards have gained traction, such as in constrained problems, exploration, and acting upon prior experiences. In this paper, we consider policy optimization in Markov Decision Problems, where the objective is a general concave utility function of the state-action occupancy measure, which subsumes several of the aforementioned examples as special cases. Such generality invalidates the Bellman equation. As this means that dynamic programming no longer works, we focus on direct policy search. Analogously to the Policy Gradient Theorem Sutton et al. (2000) available for RL with cumulative rewards, we derive a new Variational Policy Gradient Theorem for RL with general utilities, which establishes that the parametrized policy gradient may be obtained as the solution of a stochastic saddle point problem involving the Fenchel dual of the utility function. We develop a variational Monte Carlo gradient estimation algorithm to compute the policy gradient based on sample paths. We prove that the variational policy gradient scheme converges globally to the optimal policy for the general objective, though the optimization problem is nonconvex. We also establish its rate of convergence of the order O(1/t) by exploiting the hidden convexity of the problem, and proves that it converges exponentially when the problem admits hidden strong convexity. Our analysis applies to the standard RL problem with cumulative rewards as a special case, in which case our result improves the available convergence rate. Copyright © 2020, The Authors. All rights reserved.

关键词： Dynamic programming

learning-to-learn stochastic gradient descent with biased regularization

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Denevi, Giulia Ciliberto, Carlo Grazzi, Riccardo Pontil, Massimiliano Computational Statistics and Machine Learning Istituto Italiano di Tecnologia Genoa16163 Italy Department of Mathematics University of Genoa Genoa16146 Italy Department of Electrical and Electronic Engineering Imperial College of London LondonSW7 1AL United Kingdom Department of Computer Science University College London LondonWC1E 6BT United Kingdom

We study the problem of learning-to-learn: inferring a learning algorithm that works well on tasks sampled from an unknown distribution. As class of algorithms we consider Stochastic Gradient Descent on the true risk regularized by the square euclidean distance to a bias vector. We present an average excess risk bound for such a learning algorithm. This result quantifies the potential benefit of using a bias vector with respect to the unbiased case. We then address the problem of estimating the bias from a sequence of tasks. We propose a meta-algorithm which incrementally updates the bias, as new tasks are observed. The low space and time complexity of this approach makes it appealing in practice. We provide guarantees on the learning ability of the meta-algorithm. A key feature of our results is that, when the number of tasks grows and their variance is relatively small, our learning-to-learn approach has a significant advantage over learning each task in isolation by Stochastic Gradient Descent without a bias term. We report on numerical experiments which demonstrate the effectiveness of our approach. Copyright © 2019, The Authors. All rights reserved.

关键词： learning algorithms

Uncertainty-aware Pseudo-label Selection for Positive-Unlabeled learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Dorigatti, Emilio Goschenhofer, Jann Schubert, Benjamin Rezaei, Mina Bischl, Bernd Department of Statistics Ludwig-Maximilians-Universität München München80539 Germany Institute of Computational Biology Helmholtz Zentrum München German Research Center for Environmental Health Neuherberg85764 Germany Munich Center for Machine Learning München Germany Fraunhofer Institute for Integrated Circuits IIS Erlangen91058 Germany Department of Mathematics Technical University of Munich Garching bei München85748 Germany

Positive-unlabeled learning (PUL) aims at learning a binary classifier from only positive and unlabeled training data. Even though real-world applications often involve imbalanced datasets where the majority of examples belong to one class, most contemporary approaches to PUL do not investigate performance in this setting, thus severely limiting their applicability in practice. In this work, we thus propose to tackle the issues of imbalanced datasets and model calibration in a PUL setting through an uncertainty-aware pseudo-labeling procedure (PUUPL): by boosting the signal from the minority class, pseudo-labeling expands the labeled dataset with new samples from the unlabeled set, while explicit uncertainty quantification prevents the emergence of harmful confirmation bias leading to increased predictive performance. Within a series of experiments, PUUPL yields substantial performance gains in highly imbalanced settings while also showing strong performance in balanced PU scenarios across recent baselines. We furthermore provide ablations and sensitivity analyses to shed light on PUUPL's several ingredients. Finally, a real-world application with an imbalanced dataset confirms the advantage of our approach. © 2022, CC BY-NC-SA.

关键词： Sensitivity analysis

学校读者我要写书评

暂无评论

Stream C sessions 8

Stream C sessions

8th International Congress on Environmental Modelling and Software - Environmental Modelling and Software for Supporting a Sustainable Future, iEMSs 2016

作者： Sànchez-Marrè, Miquel Gibert, Karina Erechtchoukova, Marina Computational Sciences Department Universitat Politècnica de Catalunya-BarcelonaTech Campus Nord Ed. Omega. Jordi Girona 1-3 Barcelona Catalonia08034 Spain Statistics and Operations Research Department Universitat Politecnica de Catalunya - Barcelona Tech Campus Nord Ed C5. Jordi Girona 1-3 Barcelona08034 Spain Knowledge Engineering and Machine Learning Group Universitat Politecnica de Catalunya - Barcelona Tech Campus Nord Ed Omega. Jordi Girona 1-3 Barcelona08034 Spain School of Information Technology Faculty of Liberal Arts and Professional Studies York University 4700 Keele Street TorontoONM3J 1P3 Canada

Session C1: Data mining for environmental sciences session 8

学校读者我要写书评

暂无评论

Session C1: Data mining for environmental sciences session

8th International Congress on Environmental Modelling and Software - Environmental Modelling and Software for Supporting a Sustainable Future, iEMSs 2016

作者： Gibert, Karina Sànchez-Marrè, Miquel Izquierdo, Joaquín Rodríguez-Roda, Ignasi Hamilton, Serena Athanasiadis, Ioannis Ciampi, Antonio Holmes, Geoff Kuentz-Simonet, Vanessa Rambonilaza, Tina Statistics and Operations Research Department Universitat Politecnica de Catalunya Barcelona Catalonia Spain Computational Sciences Department Universitat Politècnica de Catalunya-BarcelonaTech Barcelona Catalonia Spain Knowledge Engineering and Machine Learning Group Universitat Politecnica de Catalunya - Barcelona Catalonia Spain Grupo Multidisciplinar de Modelación de Fluidos Universidad Politécnica de Valencia Spain Catalan Institute for Water Research Girona Catalonia Spain Australian National University Canberra Australia Information Technologies Group Wageningen University Netherlands McGill University Montreal Canada Waikato University New Zealand IRSTEA Institut National de Recherché en Sciences et Technologies Pour L'environnement et L'agriculture Bordeaux France

DeePN2: A deep learning-based non-Newtonian hydrodynamic model

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Fang, Lidong Ge, Pei Zhang, Lei Weinan, E. Lei, Huan Department of Computational Mathematics Science and Engineering Michigan State University MI48824 United States School of Mathematical Sciences Institute of Natural Sciences and MOE-LSC Shanghai Jiao Tong University 800 Dongchuan Road Shanghai200240 China Center for Machine Learning Research School of Mathematical Sciences Peking University Beijing100871 China AI for Science Institute Beijing100080 China Department of Mathematics and Program in Applied and Computational Mathematics Princeton University NJ08544 United States Department of Statistics and Probability Michigan State University MI48824 United States

A long standing problem in the modeling of non-Newtonian hydrodynamics of polymeric flows is the availability of reliable and interpretable hydrodynamic models that faithfully encode the underlying micro-scale polymer dynamics. The main complication arises from the long polymer relaxation time, the complex molecular structure and heterogeneous interaction. DeePN2, a deep learning-based non-Newtonian hydrodynamic model, has been proposed and has shown some success in systematically passing the micro-scale structural mechanics information to the macro-scale hydrodynamics for suspensions with simple polymer conformation and bond potential. The model retains a multi-scaled nature by mapping the polymer configurations into a set of symmetry-preserving macro-scale features. The extended constitutive laws for these macro-scale features can be directly learned from the kinetics of their micro-scale counterparts. In this paper, we develop DeePN2 using more complex micro-structural models. We show that DeePN2 can faithfully capture the broadly overlooked viscoelastic differences arising from the specific molecular structural mechanics without human intervention. © 2021, CC BY.

关键词： Non Newtonian flow