Man vs. Machine: Practical Adversarial Detection of Malicious Crowdsourcing Workers

By G. Wang, Tianyi Wang, Haitao Zheng, Ben Y. Zhao. 2014

187 citations recorded by the source corpus.

Subfields: AI Ethics, Fairness & Society

How researchers see this paper

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

In AI Ethics, Fairness & Society

Citations in the corpus

Yes, Machine Learning Can Be More Secure! A Case Study on Android Malware Detection (2017, neutral): “quently cause misclassification errors [7], [11], [25], [32], [34], [42], [43], [44], [45].”
Support vector machines under adversarial label contamination (2015, background): “Keywords: Support Vector Machines, Adversarial Learning, Label Noise, Label Flip Attacks”
Towards Making Systems Forget with Machine Unlearning (2015, methodology): “1In this paper, we use the term pollute [56] instead of poison [47, 77].”
DeepTest: Automated Testing of Deep-Neural-Network-Driven Autonomous Cars (2017, background): “Several prior works have explored defenses against these attacks with different effectiveness [29, 32, 35, 38, 41, 43, 48, 57, 64, 68, 74, 77, 80, 84, 86].”
Towards Practical Verification of Machine Learning: The Case of Computer Vision Systems (2017, background): “Adversarial machine learning is a popular research area that focuses on generating errorinducing test inputs for ML models by adding minimal perturbations to an existing input [16], [26], [27], [30], [35], [41], [44], [51], [53] and studying how to improve the robustness of the ML systems against su”
Efficient Repair of Polluted Machine Learning Systems via Causal Unlearning (2018, background): “Wang et al. [45] show that crafted training samples can mislead the machine learning classifier detecting malicious crowdsourcing workers.”
Automated Crowdturfing Attacks and Defenses in Online Review Systems (2017, background): “using human writers, workers tend to rapidly generate the requested reviews, producing a burst in new reviews that is easily detected [30, 73].”
Strength in Numbers: Robust Tamper Detection in Crowd Computations (2015, background): “A large body of work applied machine learning techniques to distinguish between behaviors (activities and proﬁle characteristics) of Sybil and non-Sybil identities [12,26,37,38].”
Auror: defending against poisoning attacks in collaborative deep learning systems (2016, background): “For example, attackers can modify the test data to avoid the detection of malicious samples on a well-trained model [13, 16, 21, 33, 34, 41, 46] or alter the training data to affect the model accuracy during the learning phase [17, 36, 42].”
CrowdTarget: Target-based Detection of Crowdturfing in Online Social Networks (2015, background): “However, recent studies [29, 32] show that applying the techniques to detect crowdturﬁng accounts is vulnerable to simple evasion techniques, such as performing malicious tasks while doing normal behaviors.”