Security and Privacy Issues in Deep Learning

By Ho Bae, Jaehee Jang, Dahuin Jung, Hyemi Jang, Heonseok Ha, Sungroh Yoon. 2018

90 citations recorded by the source corpus.

With the development of machine learning (ML), expectations for artificial intelligence (AI) technology have been increasing daily. In particular, deep neural networks have shown outstanding performance results in many fields. Many applications are deeply involved in our daily life, such as making significant decisions in application areas based on predictions or classifications, in which a DL model could be relevant. Hence, if a DL model causes mispredictions or misclassifications due to malicious external influences, then it can cause very large difficulties in real life. Moreover, training DL models involve an enormous amount of data and the training data often include sensitive information. Therefore, DL models should not expose the privacy of such data. In this paper, we review the vulnerabilities and the developed defense methods on the security of the models and data privacy under the notion of secure and private AI (SPAI). We also discuss current challenges and open issues.

Subfields: Artificial Intelligence

How researchers see this paper

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

In Artificial Intelligence

1. Cong Xie
2. Oluwasanmi Koyejo
3. Indranil Gupta
4. Li Fei-Fei
5. Peter Stone
6. Bo Liu
7. Serena Yeung
8. Sergey Levine

Citations in the corpus

• Practical Distributed Learning: Secure Machine Learning with Communication-Efficient Local Updates (2019, background): “Beyond more benign issues such as hardware, software or communication failures, there are multiple ways for nefarious users to manipulate the uploaded models e.g. data poisoning (Bae et al., 2018).”
• Continual Learning and Private Unlearning (2022, background): “From time to time, the company may be asked to expunge previously learned behaviors and/or knowledge about specific tasks that are found to raise potential fairness (Mehrabi et al., 2021), privacy or security issues (Bae et al., 2018).”
• Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inference (2018, background): “Machine Learning as a Service (MLaaS) [8] is a framework where cloud providers offer machine learning training of inference hosted on the cloud.”