Visible to the public Uncertainty-Aware Opinion Inference Under Adversarial Attacks

TitleUncertainty-Aware Opinion Inference Under Adversarial Attacks
Publication TypeConference Paper
Year of Publication2019
AuthorsAlim, Adil, Zhao, Xujiang, Cho, Jin-Hee, Chen, Feng
Conference Name2019 IEEE International Conference on Big Data (Big Data)
Date Publisheddec
ISBN Number978-1-7281-0858-2
KeywordsAdv-COI, adversarial collective opinion inference, adversarial evidence, black box adversarial attacks, collective subjective logic, composability, compositionality, computational complexity, CSL, data mining, data mining tasks, Data models, decision making, highly scalable opinion inference, inference mechanisms, learning (artificial intelligence), logic-rule based structured data, Metrics, minimax techniques, possible adversarial attacks, Probabilistic logic, probabilistic model, probabilistic soft logic, probability, Probability density function, PSL, pubcrawl, resilience, Resiliency, Robustness, security of data, Silicon, Training, uncertain evidence, Uncertainty, uncertainty-aware opinion inference, unknown opinions, white box adversarial attacks, White Box Security

Inference of unknown opinions with uncertain, adversarial (e.g., incorrect or conflicting) evidence in large datasets is not a trivial task. Without proper handling, it can easily mislead decision making in data mining tasks. In this work, we propose a highly scalable opinion inference probabilistic model, namely Adversarial Collective Opinion Inference (Adv-COI), which provides a solution to infer unknown opinions with high scalability and robustness under the presence of uncertain, adversarial evidence by enhancing Collective Subjective Logic (CSL) which is developed by combining SL and Probabilistic Soft Logic (PSL). The key idea behind the Adv-COI is to learn a model of robust ways against uncertain, adversarial evidence which is formulated as a min-max problem. We validate the out-performance of the Adv-COI compared to baseline models and its competitive counterparts under possible adversarial attacks on the logic-rule based structured data and white and black box adversarial attacks under both clean and perturbed semi-synthetic and real-world datasets in three real world applications. The results show that the Adv-COI generates the lowest mean absolute error in the expected truth probability while producing the lowest running time among all.

Citation Keyalim_uncertainty-aware_2019