检索结果-内蒙古大学图书馆

Sharp statistical guarantees for adversarially robust gaussian classification

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Dan, Chen Wei, Yuting Ravikumar, Pradeep Computer Science Department Department of Statistics and Data Science Machine Learning Department Carnegie Mellon University

Adversarial robustness has become a fundamental requirement in modern machine learning applications. Yet, there has been surprisingly little statistical understanding so far. In this paper, we provide the first result of the optimal minimax guarantees for the excess risk for adversarially robust classification, under Gaussian mixture model proposed by Schmidt et al. (2018). The results are stated in terms of the Adversarial Signal-to-Noise Ratio (AdvSNR), which generalizes a similar notion for standard linear classification to the adversarial setting. For the Gaussian mixtures with AdvSNR value of r, we establish an excess risk lower bound of order Θ(e−(1 8 +o(1))r2nd ) and design a computationally efficient estimator that achieves this optimal rate. Our results built upon minimal set of assumptions while cover a wide spectrum of adversarial perturbations including `p balls for any p ≥ 1. Copyright © 2020, The Authors. All rights reserved.

关键词： Signal to noise ratio

Position: Bayesian Deep learning is Needed in the Age of Large-Scale AI 41

学校读者我要写书评

暂无评论

Position: Bayesian Deep Learning is Needed in the Age of Lar...

41st International Conference on machine learning, ICML 2024

作者： Papamarkou, Theodore Skoularidou, Maria Palla, Konstantina Aitchison, Laurence Arbel, Julyan Dunson, David Filippone, Maurizio Fortuin, Vincent Hennig, Philipp Hernández-Lobato, José Miguel Hubin, Aliaksandr Immer, Alexander Karaletsos, Theofanis Khan, Mohammad Emtiyaz Kristiadi, Agustinus Li, Yingzhen Mandt, Stephan Nemeth, Christopher Osborne, Michael A. Rudner, Tim G.J. Rügamer, David Teh, Yee Whye Welling, Max Wilson, Andrew Gordon Zhang, Ruqi Department of Mathematics The University of Manchester Manchester United Kingdom Eric and Wendy Schmidt Center Broad Institute of MIT and Harvard Cambridge United States Spotify London United Kingdom Computational Neuroscience Unit University of Bristol Bristol United Kingdom Centre Inria de l'Université Grenoble Alpes Grenoble France Department of Statistical Science Duke University United States Statistics Program KAUST Saudi Arabia Helmholtz AI Munich Germany Department of Computer Science Technical University of Munich Munich Germany Munich Center for Machine Learning Munich Germany Tübingen AI Center University of Tübingen Tübingen Germany Department of Engineering University of Cambridge Cambridge United Kingdom Department of Mathematics University of Oslo Oslo Norway Bioinformatics and Applied Statistics Norwegian University of Life Sciences Ås Norway Department of Computer Science ETH Zurich Switzerland Chan Zuckerberg Initiative CA United States Center for Advanced Intelligence Project RIKEN Tokyo Japan Vector Institute Toronto Canada Department of Computing Imperial College London London United Kingdom Department of Computer Science UC Irvine Irvine United States Department of Mathematics and Statistics Lancaster University Lancaster United Kingdom Department of Engineering Science University of Oxford Oxford United Kingdom Center for Data Science New York University New York United States Department of Statistics LMU Munich Munich Germany DeepMind London United Kingdom Department of Statistics University of Oxford Oxford United Kingdom Informatics Institute University of Amsterdam Amsterdam Netherlands Courant Institute of Mathematical Sciences Center for Data Science Computer Science Department New York University New York United States Department of Computer Science Purdue University West Lafayette United States

In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language datasets. However, a broader perspective reveals a multitude of overlooked metrics, tasks, and data types, such as uncertainty, active and continual learning, and scientific data, that demand attention. Bayesian deep learning (BDL) constitutes a promising avenue, offering advantages across these diverse settings. This paper posits that BDL can elevate the capabilities of deep learning. It revisits the strengths of BDL, acknowledges existing challenges, and highlights some exciting research avenues aimed at addressing these obstacles. Looking ahead, the discussion focuses on possible ways to combine large-scale foundation models with BDL to unlock their full potential. Copyright 2024 by the author(s)

关键词： Contrastive learning

TNCR: Table net detection and classification dataset

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Abdallah, Abdelrahman Berendeyev, Alexander Nuradin, Islam Nurseitov, Daniyar Department of Machine Learning & Data Science Satbayev University Almaty Almaty050013 Kazakhstan National Open Research Laboratory for Information and Space Technologies Satbayev University Almaty Almaty050013 Kazakhstan

We present TNCR, a new table dataset with varying image quality collected from free websites. TNCR dataset can be used for table detection in scanned document images and their classification into 5 different classes. TNCR contains 9428 high-quality labeled images. In this paper, we have implemented state-of-the-art deep learning-based methods for table detection to create several strong baselines. Cascade Mask R-CNN with ResNeXt-101-64x4d Backbone Network achieves the highest performance compared to other methods with a precision of 79.7%, recall of 89.8%, and f1 score of 84.4% on the TNCR dataset. We have made TNCR open source in the hope of encouraging more deep learning approaches to table detection, classification and structure recognition. The dataset and trained model checkpoints are available at https://***/abdoelsayed2016/TNCR_dataset © 2021, CC BY.

关键词： Convolutional neural networks

Predicting the Initial Conditions of the Universe using a Deterministic Neural Network

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jindal, Vaibhav Liang, Albert Singh, Aarti Ho, Shirley Jamieson, Drew Machine Learning Department Carnegie Mellon University United States Cohesity Two Sigma Investments United States Center for Computational Astrophysics Flatiron Institute United States Department of Physics Center for Data Science New York University United States Department of Astrophysical Sciences Princeton University United States Max Planck Institute for Astrophysics Germany Carnegie Mellon University United States

Finding the initial conditions that led to the current state of the universe is challenging because it involves searching over an intractable input space of initial conditions, along with modeling their evolution via tools such as N-body simulations which are computationally expensive. Recently, deep learning has emerged as a surrogate for N-body simulations by directly learning the mapping between the linear input of an N-body simulation and the final nonlinear output from the simulation, significantly accelerating the forward modeling. However, this still does not reduce the search space for initial conditions. In this work, we pioneer the use of a deterministic convolutional neural network for learning the reverse mapping and show that it accurately recovers the initial linear displacement field over a wide range of scales (−1h), despite the one-to-many mapping of the inverse problem (due to the divergent backward trajectories at smaller scales). Specifically, we train a V-Net architecture, which outputs the linear displacement of an N-body simulation, given the nonlinear displacement at redshift z = 0 and the cosmological parameters. The results of our method suggest that a simple deterministic neural network is sufficient for accurately approximating the initial linear states, potentially obviating the need for the more complex and computationally demanding backward modeling methods that were recently proposed. © 2023, CC BY.

关键词： Inverse problems

Adaptive hybrid density functionals

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Khan, Danish Price, Alastair James Arthur Ach, Maximilian L. von Lilienfeld, O. Anatole Trottier, Olivier Chemical Physics Theory Group Department of Chemistry University of Toronto St. George Campus TorontoON Canada Vector Institute for Artificial Intelligence TorontoON Canada Acceleration Consortium University of Toronto TorontoON Canada Department of Physics University of Toronto St. George Campus TorontoON Canada Department of Physics Ludwig-Maximilians-Universität Munich Germany Department of Materials Science and Engineering University of Toronto St. George Campus TorontoON Canada Machine Learning Group Technische Universität Berlin Berlin10587 Germany Berlin Institute for the Foundations of Learning and Data Berlin10587 Germany

Exact exchange and correlation contributions are known to crucially affect electronic states, which in turn govern covalent bond formation and breaking in chemical species. Empirically averaging the exact exchange admixture and the amount of correlation over configurational and compositional degrees of freedom, hybrid density functional approximations (DFAs) have been widely successful, yet have fallen short to reach explicitly correlated high level quantum chemistry accuracy in general. Using quantum machine learning, we have adaptified hybrid DFAs by generating optimal admixture ratios of exact exchange or of correlation "on the fly", specifically for any chemical compound. Adaptive exchange in the PBE0 hybrid DFA yields atomization energies sufficiently accurate to effectively cure the infamous spin gap problem in DFT, and also improves electron densities, and HOMO-LUMO gaps in organic molecules. Using adaptive PBE0, we revised the entire QM9 data set presenting more accurate quantum properties that include, on average, stronger covalent binding, larger band-gaps, more localized electron densities, and slightly larger dipole-moments. Adaptive PBE correlation (in conjunction with 100% HF exchange) also affords much improved atomization energies, and even enables covalent bond dissociation. © 2024, CC BY.

关键词： Carrier concentration

APPROXIMATION THEORY, COMPUTING, AND DEEP learning ON THE WASSERSTEIN SPACE

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Fornasier, Massimo Heid, Pascal Sodini, Giacomo Enrico TUM School of Computation Information Technlogy Department of Mathematics Boltzmannstrasse 3 Garching bei München85748 Germany TUM Institute for Advanced Studies Germany Munich Data Science Institute Germany Munich Center for Machine Learning Germany Institut für Mathematik Fakultät für Mathematik Universität Wien Oskar-Morgenstern-Platz 1 Wien1090 Austria

The challenge of approximating functions in infinite-dimensional spaces from finite samples is widely regarded as formidable. In this study, we delve into the challenging problem of the numerical approximation of Sobolev-smooth functions defined on probability spaces. Our particular focus centers on the Wasserstein distance function, which serves as a relevant example. In contrast to the existing body of literature focused on approximating efficiently pointwise evaluations, we chart a new course to define functional approximants by adopting three machine learning-based approaches: 1. Solving a finite number of optimal transport problems and computing the corresponding Wasserstein potentials. 2. Employing empirical risk minimization with Tikhonov regularization in Wasserstein Sobolev spaces. 3. Addressing the problem through the saddle point formulation that characterizes the weak form of the Tikhonov functional’s Euler-Lagrange equation. As a theoretical contribution, we furnish explicit and quantitative bounds on generalization errors for each of these solutions. In the proofs, we leverage the theory of metric Sobolev spaces and we combine it with techniques of optimal transport, variational calculus, and large deviation bounds. In our numerical implementation, we harness appropriately designed neural networks to serve as basis functions. These networks undergo training using diverse methodologies. This approach allows us to obtain approximating functions that can be rapidly evaluated after training. Consequently, our constructive solutions significantly enhance at equal accuracy the evaluation speed, surpassing that of state-of-the-art methods by several orders of magnitude. This allows evaluations over large datasets several times faster, including training, than traditional optimal transport algorithms. Moreover, our analytically designed deep learning architecture slightly outperforms the test error of state-of-the-art CNN architectures on datasets of ***

关键词： Sobolev spaces

The Role of Blockchain Technology in Supply chain Management for data Security

学校读者我要写书评

暂无评论

The Role of Blockchain Technology in Supply chain Management...

Blockchain and Distributed Systems Security (ICBDS), IEEE International Conference on

作者： Dipika R. Birari Vijay U. Rathod Aarti Dandavate Nilesh Shelke Ranjit Kumar Masira M. S Kulkarni Department of Information Technology Army Institute of Technology Pune Department of Artificial Intelligent and Machine Learning G H Raisoni College of Engineering and Management Wagholi Pune Department of Computer Engineering Dhole Patil College of Engineering Pune Department of Computer Science and Engineering Symbiosis Institute of Technology Nagpur Campus Symbiosis International Deemed University Pune India Department of Computer Engineering Ajeenkya D Y Patil University Pune Department of Cyber Security and Data Science G H Raisoni College of Engineering and Management Wagholi Pune

ISBN: (数字)9798350354348

ISBN: (纸本)9798350354355

In a traditional supply chain system, the process of producing raw materials so that a product may be delivered to a customer is a manual operation with insufficient data and transaction security. The entire process is time consuming and requires a lot of effort. The undivided process is generally ineffective and unreliable for consumers. The data security, authenticity, the inclusion of will significantly improve the time management and transaction procedures of conventional supply chain management systems. data is shielded from illegal access via the ground-breaking, decentralized technology known as blockchain. After smart contracts are put into place, the complete supply chain management (SCM) will be satisfied with the customer. When the mediator is contracted, which is possible in various ways, the plan becomes more reliable. This article places particular emphasis on lowering third party involvement in the supply chain system and enhancing data security. Conventional supply chain management techniques have several serious shortcomings. Some of them include a lack of traceability, difficulties maintaining product safety and quality, a failure to track and manage inventories in stores and warehouses, escalating supply chain costs, and so on. The reduction of third-party involvement in the supply chain system and the improvement of data security are the main goals of this article. This enhances process accessibility, effectiveness, and timeliness overall. They will feel safer throughout the payment procedure, which is the main benefit. Yet in this investigation, a smart contract and a peer-to-peer encrypted system were both used. There are a few additional features as well.

关键词： Supply chain management Costs Supply chains Smart contracts Reliability engineering Raw materials Blockchains Peer-to-peer computing Product safety

Transformer-based Annotation Bias-aware Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liao, Zehui Xie, Yutong Hu, Shishuai Xia, Yong National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Northwestern Polytechnical University Xi’an710072 China Australian Institute for Machine Learning The University of Adelaide AdelaideSA Australia Ningbo Institute of Northwestern Polytechnical University Ningbo315048 China Research and Development Institute Northwestern Polytechnical University in Shenzhen Shenzhen518057 China

Manual medical image segmentation is subjective and suffers from annotator-related bias, which can be mimicked or amplified by deep learning methods. Recently, researchers have suggested that such bias is the combination of the annotator preference and stochastic error, which are modeled by convolution blocks located after decoder and pixel-wise independent Gaussian distribution, respectively. It is unlikely that convolution blocks can effectively model the varying degrees of preference at the full resolution level. Additionally, the independent pixel-wise Gaussian distribution disregards pixel correlations, leading to a discontinuous boundary. This paper proposes a Transformer-based Annotation Bias-aware (TAB) medical image segmentation model, which tackles the annotator-related bias via modeling annotator preference and stochastic errors. TAB employs the Transformer with learnable queries to extract the different preference-focused features. This enables TAB to produce segmentation with various preferences simultaneously using a single segmentation head. Moreover, TAB takes the multivariant normal distribution assumption that models pixel correlations, and learns the annotation distribution to disentangle the stochastic error. We evaluated our TAB on an OD/OC segmentation benchmark annotated by six annotators. Our results suggest that TAB outperforms existing medical image segmentation models which take into account the annotator-related bias. © 2023, CC BY-NC-ND.

关键词： Gaussian distribution

Classification under Nuisance Parameters and Generalized Label Shift in Likelihood-Free Inference

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Masserano, Luca Shen, Alex Doro, Michele Dorigo, Tommaso Izbicki, Rafael Lee, Ann B. Department of Statistics and Data Science Carnegie Mellon University Pittsburgh United States Machine Learning Department Carnegie Mellon University Pittsburgh United States Department of Physics and Astronomy Università di Padova Padova Italy Istituto Nazionale di Fisica Nucleare Sezione di Padova Italy Lulea Techniska Universitet Lulea Sweden Universal Scientific Education and Research Network Italy Department of Statistics Universidade Federal de São Carlos São Paulo Brazil

An open scientific challenge is how to classify events with reliable measures of uncertainty, when we have a mechanistic model of the data-generating process but the distribution over both labels and latent nuisance parameters is different between train and target data. We refer to this type of distributional shift as generalized label shift (GLS). Direct classification using observed data X as covariates leads to biased predictions and invalid uncertainty estimates of labels Y . We overcome these biases by proposing a new method for robust uncertainty quantification that casts classification as a hypothesis testing problem under nuisance parameters. The key idea is to estimate the classifier’s receiver operating characteristic (ROC) across the entire nuisance parameter space, which allows us to devise cutoffs that are invariant under GLS. Our method effectively endows a pretrained classifier with domain adaptation capabilities and returns valid prediction sets while maintaining high power. We demonstrate its performance on two challenging scientific problems in biology and astroparticle physics with data from realistic mechanistic models. © 2024, CC BY.

关键词：