Balazs Pejo was born in 1989 in Budapest, Hungary. He received a B.Sc. degree in Mathematics from the Budapest University of Technology and Economics (BME, Hungary) in 2012 and two M.Sc. degree in Computer Science in the Security and Privacy program of EIT Digital from the University of Trento (UNITN, Italy) and Eotvos Lorand University (ELTE, Hungary) in 2014. He earned the Ph.D. degree in Informatics from the University of Luxembourg (UNILU, Luxembourg) in 2019. Currently, he is a member of the Laboratory of Cryptography and Systems Security (CrySyS Lab).
List of Courses
- Inference Attacks
- Differential Privacy
- Machine Learning
- Game Theory
- Improving Machine Learning by Preclassification:
Machine Learning (ML) algorithm performs better on bigger datasets, so in general, it is a good idea to use more data. On the other hand, not all data was created equal: could the model's accuracy be improved by carefully selecting different training data for each phase of the learning?
- Testing Data Inference:
For every ML model, the underlying data is separated into training and testing. While Membership Inference aims to determine whether a particular data point was part of a training set, currently, there are no known techniques to indicate a data point in the test set. Is it even possible?
- Accuracy vs Privacy - Optimizing the Complexity:
More complex ML models perform better, mostly because they are capable of learning more. As a direct consequence, they could potentially leak more information than their simpler counterparts. In which situation does the accuracy gain outweigh the privacy leakage?
- Privacy-Security-Accuracy Triangle:
There is a clear connection between privacy and accuracy within ML. However, more privacy (e.g., noise) could decrease the robustness of the model as it would be easier to fool it (e.g., misclassification). Could this trade-off be measured, and based on some incentives optimized?
- Privacy-Honesty-Accuracy Trade-off:
Privacy protection has an explicit effect on accuracy (e.g., more noise, less accurate model). With more privacy comes a higher chance of cheating (since the actual contribution is more and more hidden). Hence there is an implicit effect as well. How could this relationship be modeled?
List of Students
- Mathias Parisot (BSc, VU Amsterdam): Property Inference vs Model Comlexity
- Andras Totth (BSc, BME): Distributed Approximation of the Shapley Value
- Nikolett Kapui (BSc, BME): Automatization of SOC Tasks via Machine Learning
- [2020-]: Privacy Enhancing Technologies Symposium (PETS)
- [2020-]: Workshop on Privacy in Natural Language Processing (PrivateNLP)
- [2021-]: International Conference on Emerging Security Information, Systems and Technologies (SECUWARE)
List of Publications
- Johannes Mueller, Balazs Pejo, Ivan Pryvalov and Najmeh Soroush: "A New Technique for Deniable Vote Updating: Intuitive, Efficient, and Secure", (Under Review). 2021.
- Andras Instvan Seres, Balazs Pejo, and Peter Burcsi: "Why Fuzzy Message Detection Leads to Fuzzy Privacy Guarantees", (Under Review). 2021.
- Frederick Ayala-Gomez, Ismo Horppu, Erlin Gulbenkoglu, Vesa Siivola and Balazs Pejo: "Revenue Attribution on iOS 14 using Conversion Values in F2P Games", (Under Review). 2021.
- Balazs Pejo and Gergely Biczok: "Quality Inference in Federated Learning with Secure Aggregation", (Under Review). 2021.
- Balazs Pejo, Gergely Biczok and Gergely Acs: "Measuring Contributions in Privacy-Preserving Federated Learning", ERCIM NEWS 35: Privacy Preserving Computation. ERCIM, 2021.
- Mathias Parisot, Balazs Pejo and Dayana Spagnuelo: "Property Inference Attacks on Convolutional Neural Networks: Influence and Implications of Target Model's Complexity", Proceedings on the 18th International Conference on Security and Cryptography (SECRYPT). ScitePress, 2021. [Best Poster Award]
- Balazs Pejo and Gergely Biczok: "Corona Games: Masks, Social Distancing, and Mechanism Design", Proceedings of the 1st ACM SIGSPATIAL International Workshop on Modeling and Understanding the Spread of COVID-19 (COVID). ACM, 2020.
- Damien Desfontaines and Balazs Pejo: "SoK: Differential Privacies", Proceedings on the 20th Privacy Enhancing Technologies (PoPETs). De Gruyter, 2020.
- Balazs Pejo, Qiang Tang and Gergely Biczok: "Together or Alone: The Price of Privacy in Collaborative Learning", Proceedings on the 19th Privacy Enhancing Technologies (PoPETs). De Gruyter, 2019.
- Balazs Pejo and Qiang Tang: "To Cheat or Not to Cheat: A Game-Theoretic Analysis of Outsourced Computation Verification", Proceedings of the 5th ACM International Workshop on Security in Cloud Computing (SCC). ACM, 2017.
- Qiang Tang, Balazs Pejo and Husen Wang: "Protect both integrity and confidentiality in outsourcing collaborative filtering computations", 9th International Conference on Cloud Computing (CLOUD). IEEE, 2016.