arXiv subjects

Scott Alfeld

Publications and source records attributed to Scott Alfeld.

15 recordsLinked to original sources

arXiv2026

Identifying Adversary Characteristics from an Observed Attack

When used in automated decision-making systems, machine learning (ML) models are vulnerable to data-manipulation attacks. Some defense mechanisms (e.g., adversarial regularization) directly affect the ML models while others (e.g., anomaly detection) act within the broader system. In this paper we consider a different task for defending the adversary, focusing on the attacker, rather than the attack. We present and demonstrate a framework for identifying characteristics about the attacker from an observed attack. We prove that, without additional knowledge, the attacker is non-identifiable (multiple potential attackers would perform the same observed attack). To address this challenge, we propose a domain-agnostic framework to identify the most probable attacker. This framework aids the defender in two ways. First, knowledge about the attacker can be leveraged for exogenous mitigation (i.e., addressing the vulnerability by altering the decision-making system outside the learning algorithm and/or limiting the attacker's capability). Second, when implementing defense methods that directly affect the learning process (e.g., adversarial regularization), knowledge of the specific attacker improves performance. We present the details of our framework and illustrate its applicability through specific instantiations on a variety of learners.

cs.LG↗

arXiv2023

Defense Against Shortest Path Attacks

Identifying shortest paths between nodes in a network is an important task in many applications. Recent work has shown that a malicious actor can manipulate a graph to make traffic between two nodes of interest follow their target path. In this paper, we develop a defense against such attacks by modifying the edge weights that users observe. The defender must balance inhibiting the attacker against any negative effects on benign users. Specifically, the defender's goals are: (a) recommend the shortest paths to users, (b) make the lengths of the shortest paths in the published graph close to those of the same paths in the true graph, and (c) minimize the probability of an attack. We formulate the defense as a Stackelberg game in which the defender is the leader and the attacker is the follower. We also consider a zero-sum version of the game in which the defender's goal is to minimize cost while achieving the minimum possible attack probability. We show that the defense problem is NP-hard and propose heuristic solutions for both the zero-sum and non-zero-sum settings. By relaxing some constraints of the original problem, we formulate a linear program for local optimization around a feasible point. We present defense results with both synthetic and real networks and show that our methods often reach the lower bound of the defender's cost.

cs.SI↗

arXiv2022

A Visual Active Search Framework for Geospatial Exploration

Many problems can be viewed as forms of geospatial search aided by aerial imagery, with examples ranging from detecting poaching activity to human trafficking. We model this class of problems in a visual active search (VAS) framework, which has three key inputs: (1) an image of the entire search area, which is subdivided into regions, (2) a local search function, which determines whether a previously unseen object class is present in a given region, and (3) a fixed search budget, which limits the number of times the local search function can be evaluated. The goal is to maximize the number of objects found within the search budget. We propose a reinforcement learning approach for VAS that learns a meta-search policy from a collection of fully annotated search tasks. This meta-search policy is then used to dynamically search for a novel target-object class, leveraging the outcome of any previous queries to determine where to query next. Through extensive experiments on several large-scale satellite imagery datasets, we show that the proposed approach significantly outperforms several strong baselines. We also propose novel domain adaptation techniques that improve the policy at decision time when there is a significant domain gap with the training data. Code is publicly available.

cs.CV↗

arXiv2022

Attacking Shortest Paths by Cutting Edges

Identifying shortest paths between nodes in a network is a common graph analysis problem that is important for many applications involving routing of resources. An adversary that can manipulate the graph structure could alter traffic patterns to gain some benefit (e.g., make more money by directing traffic to a toll road). This paper presents the Force Path Cut problem, in which an adversary removes edges from a graph to make a particular path the shortest between its terminal nodes. We prove that this problem is APX-hard, but introduce PATHATTACK, a polynomial-time approximation algorithm that guarantees a solution within a logarithmic factor of the optimal value. In addition, we introduce the Force Edge Cut and Force Node Cut problems, in which the adversary targets a particular edge or node, respectively, rather than an entire path. We derive a nonconvex optimization formulation for these problems, and derive a heuristic algorithm that uses PATHATTACK as a subroutine. We demonstrate all of these algorithms on a diverse set of real and synthetic networks, illustrating the network types that benefit most from the proposed algorithms.

cs.SI↗

arXiv2022

Training-Time Attacks against k-Nearest Neighbors

Nearest neighbor-based methods are commonly used for classification tasks and as subroutines of other data-analysis methods. An attacker with the capability of inserting their own data points into the training set can manipulate the inferred nearest neighbor structure. We distill this goal to the task of performing a training-set data insertion attack against $k$-Nearest Neighbor classification ($k$NN). We prove that computing an optimal training-time (a.k.a. poisoning) attack against $k$NN classification is NP-Hard, even when $k = 1$ and the attacker can insert only a single data point. We provide an anytime algorithm to perform such an attack, and a greedy algorithm for general $k$ and attacker budget. We provide theoretical bounds and empirically demonstrate the effectiveness and practicality of our methods on synthetic and real-world datasets. Empirically, we find that $k$NN is vulnerable in practice and that dimensionality reduction is an effective defense. We conclude with a discussion of open problems illuminated by our analysis.

cs.LG↗

arXiv2022

Approximate Data Deletion in Generative Models

Users have the right to have their data deleted by third-party learned systems, as codified by recent legislation such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA). Such data deletion can be accomplished by full re-training, but this incurs a high computational cost for modern machine learning models. To avoid this cost, many approximate data deletion methods have been developed for supervised learning. Unsupervised learning, in contrast, remains largely an open problem when it comes to (approximate or exact) efficient data deletion. In this paper, we propose a density-ratio-based framework for generative models. Using this framework, we introduce a fast method for approximate data deletion and a statistical test for estimating whether or not training points have been deleted. We provide theoretical guarantees under various learner assumptions and empirically demonstrate our methods across a variety of generative methods.

cs.LG↗

arXiv2022

Hard to Forget: Poisoning Attacks on Certified Machine Unlearning

The right to erasure requires removal of a user's information from data held by organizations, with rigorous interpretations extending to downstream products such as learned models. Retraining from scratch with the particular user's data omitted fully removes its influence on the resulting model, but comes with a high computational cost. Machine "unlearning" mitigates the cost incurred by full retraining: instead, models are updated incrementally, possibly only requiring retraining when approximation errors accumulate. Rapid progress has been made towards privacy guarantees on the indistinguishability of unlearned and retrained models, but current formalisms do not place practical bounds on computation. In this paper we demonstrate how an attacker can exploit this oversight, highlighting a novel attack surface introduced by machine unlearning. We consider an attacker aiming to increase the computational cost of data removal. We derive and empirically investigate a poisoning attack on certified machine unlearning where strategically designed training data triggers complete retraining when removed.

cs.LG↗

arXiv2021

Optimal Edge Weight Perturbations to Attack Shortest Paths

Finding shortest paths in a given network (e.g., a computer network or a road network) is a well-studied task with many applications. We consider this task under the presence of an adversary, who can manipulate the network by perturbing its edge weights to gain an advantage over others. Specifically, we introduce the Force Path Problem as follows. Given a network, the adversary's goal is to make a specific path the shortest by adding weights to edges in the network. The version of this problem in which the adversary can cut edges is NP-complete. However, we show that Force Path can be solved to within arbitrary numerical precision in polynomial time. We propose the PATHPERTURB algorithm, which uses constraint generation to build a set of constraints that require paths other than the adversary's target to be sufficiently long. Across a highly varied set of synthetic and real networks, we show that the optimal solution often reduces the required perturbation budget by about half when compared to a greedy baseline method.

cs.SI↗

arXiv2021

PATHATTACK: Attacking Shortest Paths in Complex Networks

Shortest paths in complex networks play key roles in many applications. Examples include routing packets in a computer network, routing traffic on a transportation network, and inferring semantic distances between concepts on the World Wide Web. An adversary with the capability to perturb the graph might make the shortest path between two nodes route traffic through advantageous portions of the graph (e.g., a toll road he owns). In this paper, we introduce the Force Path Cut problem, in which there is a specific route the adversary wants to promote by removing a minimum number of edges in the graph. We show that Force Path Cut is NP-complete, but also that it can be recast as an instance of the Weighted Set Cover problem, enabling the use of approximation algorithms. The size of the universe for the set cover problem is potentially factorial in the number of nodes. To overcome this hurdle, we propose the PATHATTACK algorithm, which via constraint generation considers only a small subset of paths -- at most 5% of the number of edges in 99% of our experiments. Across a diverse set of synthetic and real networks, the linear programming formulation of Weighted Set Cover yields the optimal solution in over 98% of cases. We also demonstrate a time/cost tradeoff using two approximation algorithms and greedy baseline methods. This work provides a foundation for addressing similar problems and expands the area of adversarial graph mining beyond recent work on node classification and embedding.

cs.SI↗

arXiv2021

RAWLSNET: Altering Bayesian Networks to Encode Rawlsian Fair Equality of Opportunity

We present RAWLSNET, a system for altering Bayesian Network (BN) models to satisfy the Rawlsian principle of fair equality of opportunity (FEO). RAWLSNET's BN models generate aspirational data distributions: data generated to reflect an ideally fair, FEO-satisfying society. FEO states that everyone with the same talent and willingness to use it should have the same chance of achieving advantageous social positions (e.g., employment), regardless of their background circumstances (e.g., socioeconomic status). Satisfying FEO requires alterations to social structures such as school assignments. Our paper describes RAWLSNET, a method which takes as input a BN representation of an FEO application and alters the BN's parameters so as to satisfy FEO when possible, and minimize deviation from FEO otherwise. We also offer guidance for applying RAWLSNET, including on recognizing proper applications of FEO. We demonstrate the use of our system with publicly available data sets. RAWLSNET's altered BNs offer the novel capability of generating aspirational data for FEO-relevant tasks. Aspirational data are free from the biases of real-world data, and thus are useful for recognizing and detecting sources of unfairness in machine learning algorithms besides biased data.

cs.CY↗

arXiv2021

Optimizing Graph Structure for Targeted Diffusion

The problem of diffusion control on networks has been extensively studied, with applications ranging from marketing to controlling infectious disease. However, in many applications, such as cybersecurity, an attacker may want to attack a targeted subgraph of a network, while limiting the impact on the rest of the network in order to remain undetected. We present a model POTION in which the principal aim is to optimize graph structure to achieve such targeted attacks. We propose an algorithm POTION-ALG for solving the model at scale, using a gradient-based approach that leverages Rayleigh quotients and pseudospectrum theory. In addition, we present a condition for certifying that a targeted subgraph is immune to such attacks. Finally, we demonstrate the effectiveness of our approach through experiments on real and synthetic networks.

cs.SI↗

arXiv2018

Training Set Camouflage

We introduce a form of steganography in the domain of machine learning which we call training set camouflage. Imagine Alice has a training set on an illicit machine learning classification task. Alice wants Bob (a machine learning system) to learn the task. However, sending either the training set or the trained model to Bob can raise suspicion if the communication is monitored. Training set camouflage allows Alice to compute a second training set on a completely different -- and seemingly benign -- classification task. By construction, sending the second training set will not raise suspicion. When Bob applies his standard (public) learning algorithm to the second training set, he approximately recovers the classifier on the original task. Training set camouflage is a novel form of steganography in machine learning. We formulate training set camouflage as a combinatorial bilevel optimization problem and propose solvers based on nonlinear programming and local search. Experiments on real classification tasks demonstrate the feasibility of such camouflage.

cs.CR↗

arXiv2018

Adversarial Regression with Multiple Learners

Despite the considerable success enjoyed by machine learning techniques in practice, numerous studies demonstrated that many approaches are vulnerable to attacks. An important class of such attacks involves adversaries changing features at test time to cause incorrect predictions. Previous investigations of this problem pit a single learner against an adversary. However, in many situations an adversary's decision is aimed at a collection of learners, rather than specifically targeted at each independently. We study the problem of adversarial linear regression with multiple learners. We approximate the resulting game by exhibiting an upper bound on learner loss functions, and show that the resulting game has a unique symmetric equilibrium. We present an algorithm for computing this equilibrium, and show through extensive experiments that equilibrium models are significantly more robust than conventional regularized linear regression.

cs.LG↗

arXiv2018

Adversarial Classification on Social Networks

The spread of unwanted or malicious content through social media has become a major challenge. Traditional examples of this include social network spam, but an important new concern is the propagation of fake news through social media. A common approach for mitigating this problem is by using standard statistical classification to distinguish malicious (e.g., fake news) instances from benign (e.g., actual news stories). However, such an approach ignores the fact that malicious instances propagate through the network, which is consequential both in quantifying consequences (e.g., fake news diffusing through the network), and capturing detection redundancy (bad content can be detected at different nodes). An additional concern is evasion attacks, whereby the generators of malicious instances modify the nature of these to escape detection. We model this problem as a Stackelberg game between the defender who is choosing parameters of the detection model, and an attacker, who is choosing both the node at which to initiate malicious spread, and the nature of malicious entities. We develop a novel bi-level programming approach for this problem, as well as a novel solution approach based on implicit function gradients, and experimentally demonstrate the advantage of our approach over alternatives which ignore network structure.

cs.MA↗

arXiv2015

Contamination Estimation via Convex Relaxations

Identifying anomalies and contamination in datasets is important in a wide variety of settings. In this paper, we describe a new technique for estimating contamination in large, discrete valued datasets. Our approach considers the normal condition of the data to be specified by a model consisting of a set of distributions. Our key contribution is in our approach to contamination estimation. Specifically, we develop a technique that identifies the minimum number of data points that must be discarded (i.e., the level of contamination) from an empirical data set in order to match the model to within a specified goodness-of-fit, controlled by a p-value. Appealing to results from large deviations theory, we show a lower bound on the level of contamination is obtained by solving a series of convex programs. Theoretical results guarantee the bound converges at a rate of $O(\sqrt{\log(p)/p})$, where p is the size of the empirical data set.

cs.IT↗