The Security of Latent Dirichlet Allocation

By Shike Mei, Xiaojin Zhu. 2015

72 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods, Privacy & Security

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Probabilistic & Bayesian Methods

In Privacy & Security

Citations in the corpus

Understanding Black-box Predictions via Influence Functions (2017, methodology): “This attack is mathematically equivalent to the gradient-based training set attacks explored by Biggio et al. (2012); Mei & Zhu (2015b) and others in the context of different models.”
Stronger data poisoning attacks break data sanitization defenses (2018, methodology): “(2012b), and has subsequently been extended to linear and logistic regression (Mei and Zhu, 2015b), topic modeling (Mei & Zhu, 2015a), collaborative filtering (Li et al., 2016), and neural networks (Koh & Liang, 2017; Yang et al., 2017; MuñozGonzález et al.”
Spectral Signatures in Backdoor Attacks (2018, background): “One well-studied setting for such training set attacks is data poisoning [3, 34, 25, 18, 31].”
Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning (2017, methodology): “After presenting the details of the threat model above, we emphasize again the differences between this work and a long line of existing work studying poisoning attacks and defenses [10], [38], [12], [11], [68], [44], [9], [35], [70], [39], [61].”
Certified Defenses for Data Poisoning Attacks (2017, methodology): “…others have studied the LASSO (Xiao et al., 2015a), clustering (Biggio et al., 2013; 2014c), PCA (Rubinstein et al., 2009), topic modeling (Mei and Zhu, 2015a), collaborative ﬁltering (Li et al., 2016), neural networks (Yang et al., 2017), and other models (Mozaffari-Kermani et al., 2015;…”
Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning (2018, methodology): “Similar techniques were later generalized to optimize data poisoning attacks for several other important learning algorithms, such as feature selection for classification [55], topic modeling [36], autoregressive models [1], collaborative filtering [33], and simple neural network architectures [39].”
Data Poisoning Attacks on Factorization-Based Collaborative Filtering (2016, methodology): “Similar techniques were later generalized to optimize data poisoning attacks for several other important learning algorithms, such as Lasso regression [7], topic modeling [8], and autoregressive models [17].”
Data Poisoning Attacks against Autoregressive Models (2016, background): “Much of the work has focused on the common settings of regression and classification, although recent work has also looked at the security of LDA (Mei and Zhu 2015a).”
Toward a general, scaleable framework for Bayesian teaching with applications to topic models (2016, background): “Mei & Zhu [12] cast teaching as bilevel optimization, which is NP-hard, and offer a solution for the limited case when the data are differentiable and the learner’s objective function is convex and regular.”
A Survey on Security Threats and Defensive Techniques of Machine Learning: A Data Driven View (2018, methodology): “In addition, the poisoning attack also threatens many widely used machine learning algorithms, e.g., SVM [43], [44], neural networks [45], Latent Dirichlet Allocation [46], principle component analysis (PCA) and LASSO [47].”
Data Poisoning Attacks on Multi-Task Relationship Learning (2018, methodology): “Inspired by (Li et al. 2016; Mei and Zhu 2015b; Xiao et al. 2015; Zhao et al. 2017), we use a projected gradient ascent method to solve our proposed bilevel problem.”
A Berkeley View of Systems Challenges for AI (2017, background): “…attacks happen at the training stage, where an adversary injects poisoned data (e.g., data with wrong labels) into the training data set that cause the learning system to learn the wrong model, such that the adversary thereby has input data in-correctly classified by the learner [73, 74, 113].”
A Theoretical Framework for Robustness of (Deep) Classifiers against Adversarial Examples (2016, background): “Multiple recent papers (Alfeld et al., 2016; Mei & Zhu, 2015b; Biggio et al., 2014; 2012; Mei & Zhu, 2015a) have considered the problem of an adversary being able to pollute the training data with the goal of inﬂuencing learning systems including support vector machines (SVM), autoregressive models…”
The Teaching Dimension of Linear Learners (2015, background): “ize these optimal lessons. Another application of optimal teaching is in computer security. In particular, optimal teaching is the mathematical formalism to study the so-called data poisoning attacks [8, 20, 19, 1]. Here the “teacher” is an attacker who has a nefarious target model in mind. The “stu”
Poison Attacks against Text Datasets with Conditional Adversarially Regularized Autoencoder (2020, background): “A predecessor of BP, called data poisoning, also poisons the training dataset of the victim model (Nelson et al., 2008; Biggio et al., 2012; Xiao et al., 2015; Mei and Zhu, 2015; Koh and Liang, 2017; Steinhardt et al., 2017) with the aim of reducing the model’s generalization.”
An Overview of Machine Teaching (2018, background): “• In one type of adversarial attack known as training-set poisoning, an attacker manipulates the behavior of a machine learning system by maliciously modifying the training data [2,30,31].”
Some Submodular Data-Poisoning Attacks on Machine Learners (2015, methodology): “Recently, we proposed a uniﬁed attack framework [14, 15] to compute the optimal data-poisoning attacks on machine learners.”
Robust Linear Regression Against Training Data Poisoning (2017, methodology): “Similar poisoning attack techniques have also been generalized to other widely used learning algorithms, such as Lasso regression [29], topic modeling [22], and autoregressive models [1].”