Filters








27,436 Hits in 3.5 sec

Verifying Robustness of Gradient Boosted Models [article]

Gil Einziger, Maayan Goldstein, Yaniv Sa'ar, Itai Segall
2019 arXiv   pre-print
This work introduces VeriGB, a tool for quantifying the robustness of gradient boosted models.  ...  Robustness to small perturbations of the input is an important quality measure for machine learning models, but the literature lacks a method to prove the robustness of gradient boosted models.  ...  The main contribution of this work is the VERIGB tool for verifying the robustness of gradient boosted models.  ... 
arXiv:1906.10991v1 fatcat:hcok4igcprgjrcqi4iqtfkdkny

Verifying Robustness of Gradient Boosted Models

Gil Einziger, Maayan Goldstein, Yaniv Sa'ar, Itai Segall
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
introduces VERIGB, a tool for quantifying the robustness of gradient boosted models.  ...  Robustness to small perturbations of the input is an important quality measure for machine learning models, but the literature lacks a method to prove the robustness of gradient boosted models.This work  ...  The main contribution of this work is the VERIGB tool for verifying the robustness of gradient boosted models.  ... 
doi:10.1609/aaai.v33i01.33012446 fatcat:dfvjf3ylc5eu7ilvxfap32cc6u

Formal Verification of Input-Output Mappings of Tree Ensembles [article]

John Törnblom, Simin Nadjm-Tehrani
2019 arXiv   pre-print
This paper presents the implementation of the method in the tool VoTE (Verifier of Tree Ensembles), and evaluates its scalability on two case studies presented in current literature.  ...  Our work also studies the limitations of the method with high-dimensional data and preliminarily investigates the trade-off between large number of trees and time taken for verification.  ...  Since the precise technique does not scale well on models trained on high-dimensional data, we were unable to verify the plausibility of range property of gradient boosting machines in this case study.  ... 
arXiv:1905.04194v1 fatcat:vwfcqjyswvc3peth6g4svdlzcq

Enhancing Certifiable Robustness via a Deep Model Ensemble [article]

Huan Zhang, Minhao Cheng, Cho-Jui Hsieh
2019 arXiv   pre-print
We propose an algorithm to enhance certified robustness of a deep model ensemble by optimally weighting each base model.  ...  RobBoost allows us to further improve certified robustness and clean accuracy by creating an ensemble of already certified models.  ...  Since there are only 5 base models, RobBoost ensembles provide a small but consistent performance advantage in verified error. Gradient Boosting of Robust Ensemble.  ... 
arXiv:1910.14655v1 fatcat:o3p5mlgpbvd7pjozf7r5t6xxxy

Learning Security Classifiers with Verified Global Robustness Properties [article]

Yizheng Chen, Shiqi Wang, Yue Qin, Xiaojing Liao, Suman Jana, David Wagner
2021 arXiv   pre-print
We structure our classifier as an ensemble of logic rules and design a new verifier to verify the properties.  ...  XGBoost model that doesn't satisfy any property.  ...  Similar to most existing robust machine learning training strategies, training a verifiably robust model is significantly slower than training a non-robust model.  ... 
arXiv:2105.11363v1 fatcat:apeyqnltarbxzhjgb57kbui4pe

Scaling up Memory-Efficient Formal Verification Tools for Tree Ensembles [article]

John Törnblom, Simin Nadjm-Tehrani
2021 arXiv   pre-print
To guarantee that machine learning models yield outputs that are not only accurate, but also robust, recent works propose formally verifying robustness properties of machine learning models.  ...  , and the second to assess the ability to verify versatile correctness properties.  ...  [7] present the tool VeriGB for verifying the robustness of gradient boosting machines. They encode the verification problem as an SMT formula, and use an SMT solver for verification.  ... 
arXiv:2105.02595v1 fatcat:pnmjlqtvwngznnbgajtw3unhya

BIGRoC: Boosting Image Generation via a Robust Classifier [article]

Roy Ganz, Michael Elad
2022 arXiv   pre-print
Our method, termed BIGRoC (Boosting Image Generation via a Robust Classifier), is based on a post-processing procedure via the guidance of a given robust classifier and without a need for additional training  ...  of the generative model.  ...  Boosting Image Generation via a Robust Classifier We propose a method for improving the quality of images synthesized by trained generative models, named BIGRoC: Boosting Image Generation via a Robust  ... 
arXiv:2108.03702v3 fatcat:4fndttdsl5gv5d32ol3fqyxhay

Multithreading AdaBoost framework for object recognition

Jinhui Chen, Tetsuya Takiguchi, Yasuo Ariki
2015 2015 IEEE International Conference on Image Processing (ICIP)  
Our research focuses on the study of effective feature description and robust classifier technique, proposing a novel learning framework, which is capable of processing multiclass objects recognition simultaneously  ...  The framework adopts rotation-invariant histograms of oriented gradients (Ri-HOG) as feature descriptors.  ...  In order to improve boosting convergence speed and accuracy, we do not use the source code of Open CV, but using the released codes of Li et al's cascade model to adopt Ri-HOG and implement our boosting  ... 
doi:10.1109/icip.2015.7350997 dblp:conf/icip/ChenTA15 fatcat:hq4z5fjpmngvvorkgapwqjyqgq

Intelligent Fault Diagnosis of Diesel Engines via Extreme Gradient Boosting and High-Accuracy Time–Frequency Information of Vibration Signals

Jianfeng Tao, Chengjin Qin, Weixing Li, Chengliang Liu
2019 Sensors  
The results verify that the highest classification accuracy of the proposed extreme gradient boosting-based algorithm is up to 99.93%.  ...  Eventually, to avoid model overfitting, the extreme gradient boosting algorithm was utilized for diesel engine misfire fault diagnosis.  ...  Conflicts of Interest: The authors declare no conflicts of interest.  ... 
doi:10.3390/s19153280 fatcat:mpyjammnvzekfiy2gzvtvlutnq

Gradient Boosting Hybridized with Exponential Natural Evolution Strategies for Estimating the Strength of Geopolymer Self-Compacting Concrete

Samuel Alves Basilio, Leonardo Goliatt
2022 Knowledge-Based Engineering and Sciences  
This study presents a gradient boosting algorithm hybridized with Natural Exponential Evolution Strategies inspired by nature to predict the mechanical properties of geopolymeric self-compacting concrete  ...  Machine learning models then emerge as surrogate models to perform this difficult task. The very design of such models has become a challenge for machine learning.  ...  Conflicts of Interest: The authors have no conflict of interest to any part. References  ... 
doi:10.51526/kbes.2022.3.1.1-16 fatcat:tcqijhxrcfc3jfv7zr3z6cfxwy

People detection in low-resolution video with non-stationary background

Jianguo Zhang, Shaogang Gong
2009 Image and Vision Computing  
Our model utilizes appearance features together with short-and long-term motion information. In particular, we boost Integral Gradient Orientation histograms of appearance and short-term motion.  ...  Experiments show that our model is more robust with better detection rate compared to the model of Viola et al.  ...  robust.  ... 
doi:10.1016/j.imavis.2008.06.013 fatcat:kiyqhijynzaapguetrn4pj4nqi

Education Data Mining Application for Predicting Students' Achievements of Portuguese Using Ensemble Model

Shuai Zhang, Jie Chen, Wenyu Zhang, Qiwei Xu, Jiaxuan Shi
2021 Science Journal of Education  
The experiments are presented for verifying the superiority of the proposed model by comparing with five base classifiers, including gradient boosting decision tree, adaptive boosting, extreme gradient  ...  Finally, two base classifiers, i.e. gradient boosting decision tree and extreme gradient boosting, are integrated to form the ensemble model.  ...  Acknowledgements The work has been supported by Zhejiang Higher Education Teaching Reform Research Project of China (No. JG20190294).  ... 
doi:10.11648/j.sjedu.20210902.16 fatcat:ggxhkbafgjayvpp2f3ui22r2fa

Binary Ensemble Neural Network: More Bits per Network or More Networks per Bit?

Shilin Zhu, Xin Dong, Hao Su
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We conclude that the error of BNNs are predominantly caused by the intrinsic instability (training time) and non-robustness (train & test time).  ...  We find that our BENN, which is faster and more robust than state-of-the-art binary networks, can even surpass the accuracy of the full-precision floating number network with the same architecture.  ...  This verifies that boosting is a better choice if the model does not overfit much.  ... 
doi:10.1109/cvpr.2019.00506 dblp:conf/cvpr/ZhuDS19 fatcat:fed4idlbqrcnzg2kicg5uoazpu

Boosting Based Conditional Quantile Estimation for Regression and Binary Classification [chapter]

Songfeng Zheng
2010 Lecture Notes in Computer Science  
Quantile Boost Regression (QBR) performs gradient descent in functional space to minimize the objective function used by quantile regression (QReg).  ...  Furthermore, QBoost is more robust to noisy predictors.  ...  Acknowledgement This work was supported by CNAS Summer Research Fellowship of Missouri State University.  ... 
doi:10.1007/978-3-642-16773-7_6 fatcat:bksyzcfxmvfuvagycxrl2l3d2i

Genetic Algorithm with SRM SVM Classifier for Face Verification

K.M Safiya
2012 International Journal of Computer Science & Information Technology (IJCSIT)  
Gradient Orientation of each color channel of human faces is robust under age progression.  ...  The problem of designing and evaluating discriminative approaches without explicit age modelling is used. To find the gradient orientation discard magnitude information.  ...  Fig 6 6 is calculated for SVM ,Boosting SVM and SRM-SVM with GA classifiers by taking the number of correctly verified individuals to the total number of images taken.  ... 
doi:10.5121/ijcsit.2012.4414 fatcat:kwm2ndzeazggzmcxv4r67skani
« Previous Showing results 1 — 15 out of 27,436 results