Learning Realistic Patterns from Visually Unrealistic Stimuli: Generalization and Data Anonymization (Extended Abstract)

Learning Realistic Patterns from Visually Unrealistic Stimuli: Generalization and Data Anonymization (Extended Abstract)

Konstantinos Nikolaidis, Stein Kristiansen, Thomas Plagemann, Vera Goebel, Knut Liestøl, Mohan Kankanhalli, Gunn-Marit Traaen, Britt Øverland, Harriet Akre, Lars Aakeroy, Sigurd Steinshamn

Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence
Journal Track. Pages 5762-5766. https://doi.org/10.24963/ijcai.2022/806

Good training data is a prerequisite to develop useful Machine Learning applications. However, in many domains existing data sets cannot be shared due to privacy regulations (e.g., from medical studies). This work investigates a simple yet unconventional approach for anonymized data synthesis to enable third parties to benefit from such anonymized data. We explore the feasibility of learning implicitly from visually unrealistic, task-relevant stimuli, which are synthesized by exciting the neurons of a trained deep neural network. As such, neuronal excitation can be used to generate synthetic stimuli. The stimuli data is used to train new classification models. Furthermore, we extend this framework to inhibit representations that are associated with specific individuals. Extensive comparative empirical investigation shows that different algorithms trained on the stimuli are able to generalize successfully on the same task as the original model.
Keywords:
Computer Vision: Representation Learning
Computer Vision: Neural generative models, auto encoders, GANs  
Natural Language Processing: Knowledge Extraction
Knowledge Representation and Reasoning: General