3,186 Hits in 3.8 sec

Perceptual Image Super-Resolution with Progressive Adversarial Network [article]

Lone Wong, Deli Zhao, Shaohua Wan, Bo Zhang
2020 arXiv   pre-print
In this paper, we argue that the curse of dimensionality is the underlying reason of limiting the performance of state-of-the-art algorithms.  ...  The key principle of PAN is that we do not apply any distance-based reconstruction errors as the loss to be optimized, thus free from the restriction of the curse of dimensionality.  ...  Curse of Dimensionality The principal argument about the issue in high-dimensional spaces is that the concept of distance-based nearest neighbors is no longer meaningful when the dimension goes sufficiently  ... 
arXiv:2003.03756v4 fatcat:dg32vyec5ndhrhmpin7kp4uhwi

Systematic Literature Review on the Anonymization of High Dimensional Streaming Datasets for Health Data Sharing

Benjamin Eze, Liam Peyton
2015 Procedia Computer Science  
Instead of static, independent tables, health data is in relational databases with multiple high-dimensional tables that are transactional and constantly changing.  ...  Relevant papers are analyzed, categorized and compared in terms of scope, and contributions.  ...  23 show that multi-dimensional data in their natural forms suffer from dimensionality curse.  ... 
doi:10.1016/j.procs.2015.08.353 fatcat:x4amox7ox5brrjzazsmk7lf6rm

The Limitations of Adversarial Training and the Blind-Spot Attack [article]

Huan Zhang, Hongge Chen, Zhao Song, Duane Boning, Inderjit S. Dhillon, Cho-Jui Hsieh
2019 arXiv   pre-print
difficult due to the curse of dimensionality and the scarcity of training data.  ...  Most importantly, for large datasets with high dimensional and complex data manifold (CIFAR, ImageNet, etc), the existence of blind-spots in adversarial training makes defending on any valid test examples  ...  Using data augmentation may eliminate some blind-spots, however for high dimensional data it is impossible to enumerate all possible inputs due to the curse of dimensionality.  ... 
arXiv:1901.04684v1 fatcat:seend3ymybec7bj5jpa2xxsmji

Template Attacks vs. Machine Learning Revisited (and the Curse of Dimensionality in Side-Channel Analysis) [chapter]

Liran Lerman, Romain Poussier, Gianluca Bontempi, Olivier Markowitch, François-Xavier Standaert
2015 Lecture Notes in Computer Science  
In this paper, we aim to contribute to the understanding of their respective strengths and weaknesses, with a particular focus on their curse of dimensionality.  ...  First and from a theoretical point of view, the data complexity of template attacks is not sensitive to the dimension increase in side-channel traces given that their profiling is perfect.  ...  Standaert is a research associate of the Belgian Fund for Scientific Research (FNRS-F.R.S.). This work has been funded in parts by the European Commission through the ERC project 280141 (CRASH).  ... 
doi:10.1007/978-3-319-21476-4_2 fatcat:6ktccfp53rcw7krwcks5xj44ze

Population Anomaly Detection through Deep Gaussianization [article]

David Tolpin
2018 arXiv   pre-print
This method is applicable to detection of 'soft' anomalies in arbitrarily distributed highly-dimensional data.  ...  A soft, or population, anomaly is characterized by a shift in the distribution of the data set, where certain elements appear with higher probability than anticipated.  ...  as 'the curse of dimensionality.'  ... 
arXiv:1805.02123v1 fatcat:tof4oqbidreallae4fwczohv4q

Adversarial Training with Voronoi Constraints [article]

Marc Khoury, Dylan Hadfield-Menell
2019 arXiv   pre-print
In particular, we highlight the importance of codimension: for low-dimensional data manifolds embedded in high-dimensional space there are many directions off the manifold in which an adversary could construct  ...  We propose a geometric framework, drawing on tools from the manifold reconstruction literature, to analyze the high-dimensional geometry of adversarial examples.  ...  In higher-dimensional embedding spaces (large d), manifold reconstruction algorithms face the curse of dimensionality.  ... 
arXiv:1905.01019v1 fatcat:fldbvjrhknfudf7vxchhqj7kqe

The Dilemma Between Data Transformations and Adversarial Robustness for Time Series Application Systems [article]

Sheila Alemany, Niki Pissinou
2021 arXiv   pre-print
Theoretical evidence has shown that the high intrinsic dimensionality of datasets facilitates an adversary's ability to develop effective adversarial examples in classification models.  ...  For example, we have seen this through dimensionality reduction techniques used to aid with the generalization of features in machine learning applications.  ...  Data dimensionality has been referred to as a "curse" due to substantial computational complexity yielding difficulties when abstracting properties in data that do not oc-cur in lower-dimensional data  ... 
arXiv:2006.10885v2 fatcat:i5zxkx3fcbeqnel42pfwjlr6aa

Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environment

Shuo Feng, Xintao Yan, Haowei Sun, Yiheng Feng, Henry X. Liu
2021 Nature Communications  
However, due to the high dimensionality of the environment and the rareness of safety-critical events, hundreds of millions of miles would be required to demonstrate the safety performance of autonomous  ...  We discover that sparse but adversarial adjustments to the naturalistic driving environment, resulting in the naturalistic and adversarial driving environment, can significantly reduce the required test  ...  Moreover, we have also provided the simulation architecture of our approach in Supplementary Fig. 1  ... 
doi:10.1038/s41467-021-21007-8 pmid:33531506 fatcat:ldwvk22vknfvndxs3jpwtwiw6a

Integrated Learning and Feature Selection for Deep Neural Networks in Multispectral Images

Anthony Ortiz, Alonso Granados, Olac Fuentes, Christopher Kiekintveld, Dalton Rosario, Zachary Bell
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
The curse of dimensionality is a well-known phenomenon that arises when applying machine learning algorithms to highly-dimensional data; it degrades performance as a function of increasing dimension.  ...  In this work, we propose an end-to-end framework to effectively integrate input feature selection into the training procedure of a deep neural network for dimensionality reduction.  ...  Adversarial examples in the physical world are normally accomplished by printing the color image of the adversarial examples.  ... 
doi:10.1109/cvprw.2018.00165 dblp:conf/cvpr/OrtizGFKRB18 fatcat:eqx7zyuzijgxhjcqjvmvhtwwuy

Partial differential equation regularization for supervised machine learning [article]

Adam M Oberman
2019 arXiv   pre-print
Implicit regularization in deep learning examples are presented, including data augmentation, adversarial training, and additive noise. These methods are reframed as explicit gradient regularization.  ...  This article is an overview of supervised machine learning problems for regression and classification.  ...  Curse of dimensionality.  ... 
arXiv:1910.01612v1 fatcat:kzinmrwycfh7fpb227q3j6viz4

Privacy protection for RFID data

Benjamin C. M. Fung, Ming Cao, Bipin C. Desai, Heng Xu
2009 Proceedings of the 2009 ACM symposium on Applied Computing - SAC '09  
RFID data by default is high-dimensional and sparse, so applying traditional K-anonymity to RFID data suffers from the curse of high dimensionality, and would result in poor data usefulness.  ...  To the best of our knowledge, this is the first work on anonymizing high-dimensional RFID data.  ...  ACKNOWLEDGEMENTS The research is supported in part by the Discovery Grants (356065-2008) from the Natural Sciences and Engineering Research Council of Canada (NSERC), and the Faculty Startup Funds from  ... 
doi:10.1145/1529282.1529626 dblp:conf/sac/FungCDX09 fatcat:qk6crb6xfze2pdmsg22f2tsyya

Service oriented architecture and privacy preserving mashup of healthcare data

R. Vijayalakshmi, N. Duraipandian
2014 International Journal of Engineering & Technology  
Mashup of health care data from different medical sources must be privacy preserved since the data recipient and/or the data provider may not always be a trusted party.  ...  LKC model overcomes this curse, by taking advantage of one of the limitations of the adversary -the adversary will not know the values of all the QID attributes.  ...  Older models such as k-anonymity, l-diversity, (α,k) anonymity, confidence bounding model etc. suffer from "curse of high-dimensionality" when used for high dimensional mashup i.e. as the number of attributes  ... 
doi:10.14419/ijet.v3i3.2139 fatcat:m5iovm4zb5gn7kqoiumkjfkj7y

Reinforcement learning for the soccer dribbling task

Arthur Carvalho, Renato Oliveira
2011 2011 IEEE Conference on Computational Intelligence and Games (CIG'11)  
We propose a reinforcement learning solution to the soccer dribbling task, a scenario in which a soccer agent has to go from the beginning to the end of a region keeping possession of the ball, as an adversary  ...  Our experiments show that, after the training period, the dribbler is able to accomplish its task against a strong adversary around 58 of the time.  ...  Glanz, and others from the Department of Electrical and Computer Engineering at the University of New Hampshire for making their CMAC code available.  ... 
doi:10.1109/cig.2011.6031994 dblp:conf/cig/CarvalhoO11 fatcat:2m4euttksvaftasgolfs6r56ly

Discriminative Multi-level Reconstruction under Compact Latent Space for One-Class Novelty Detection [article]

Jaewoo Park, Yoon Gyo Jung, Andrew Beng Jin Teoh
2021 arXiv   pre-print
Extensive experiments on public image datasets validate the effectiveness of our proposed model on both novelty and adversarial example detection, delivering state-of-the-art performance.  ...  Moreover, to properly measure the reconstruction error of high-dimensional data, a metric is required that captures high-level semantics of the data.  ...  As to the first reason, merely minimizing the L 2 distance x in −G(E(x in )) 2 is not effective due to the curse of dimensionality.  ... 
arXiv:2003.01665v3 fatcat:267bbk6ziza5zoenwtsotgnxry

Adversarial Examples Detection in Deep Networks with Convolutional Filter Statistics [article]

Xin Li, Fuxin Li
2017 arXiv   pre-print
However, recent research has shown that there exist many adversarial examples that can negatively impact the performance of such an architecture.  ...  After detecting adversarial examples, we show that many of them can be recovered by simply performing a small average filter on the image.  ...  Acknowledgements This paper was supported by Future of Life grants 2015-143880 and 2016-158701.  ... 
arXiv:1612.07767v2 fatcat:shj6kot57vchzfkxbmebxztu5m
« Previous Showing results 1 — 15 out of 3,186 results