Georg Neugebauer

Fair and privacy-preserving multi-party protocols for reconciling ordered input sets

In this paper, we introduce the first protocols for multiparty, privacy-preserving, fair reconciliation of ordered sets. Our contributions are twofold. First, we show that it is possible to extend the round-based construction for fair, two-party privacy-preserving reconciliation of ordered sets to multiple parties using a multiparty privacy-preserving set intersection protocol. Second, we propose new constructions for fair, multi-party, privacy-preserving reconciliation of ordered sets based on multiset operations. We prove that all our protocols are privacy-preserving in the semi-honest model.We furthermore provide a detailed performance analysis of our new protocols and show that the constructions based on multisets generally outperform the round-based approach.

Enabling fair and privacy-preserving applications using reconciliation protocols on ordered sets

Fair and privacy-preserving reconciliation protocols on ordered sets have been introduced recently. Despite the fact that these protocols promise to have a great impact in a variety of applications, so far their practical use has been explored to a limited extent only. This paper addresses this gap. As main contributions, this paper identifies e-voting, auctions, event scheduling, and policy reconciliation as four far-reaching areas of application and shows how fair and privacy-preserving reconciliation protocols can be used effectively in these contexts.

To model or not to model: Controlling Pac-Man ghosts without incorporating global knowledge

The creation of interesting opponents for human players in computer games is an interesting and challenging task. In contrast to up-to-date computer games, e.g. real time strategy games, learning of non-player-character strategies for older games seems to be easier and not that time-consuming. This way, older games, like the famous arcade game Pac-Man, serve as a test bed for the creation of strategies that are fun to play against. The paper at hand uses computational intelligence methods to accomplish this challenge, namely evolutionary algorithms (EA) and artificial neural networks (ANN). The latter are trained on a model of the game whereas the EA learn good behavior by playing. The performance of these two approaches is compared on the original Pac-Man level as well as on other maps with different properties to test the ability of generalizing the learned strategies.

Privacy-Preserving Multi-Party Reconciliation Secure in the Malicious Model

The problem of fair and privacy-preserving ordered set reconciliation arises in a variety of applications like auctions, e-voting, and appointment reconciliation. While several multi-party protocols have been proposed that solve this problem in the semi-honest model, there are no multi-party protocols that are secure in the malicious model so far. In this paper, we close this gap. Our newly proposed protocols are shown to be secure in the malicious model based on a variety of novel non-interactive zero-knowledge-proofs. We describe the implementation of our protocols and evaluate their performance in comparison to protocols solving the problem in the semi-honest case.

Privacy-Preserving Multi-party Reconciliation Using Fully Homomorphic Encryption

Fully homomorphic cryptosystems allow the evaluation of arbitrary Boolean circuits on encrypted inputs and therefore have very important applications in the area of secure multi-party computation. Since every computable function can be expressed as a Boolean circuit, it is theoretically clear how to achieve function evaluation on encrypted inputs. However, the transformation to Boolean circuits is not trivial in practice. In this work, we design such a transformation for certain functions, i.e., we propose algorithms and protocols which make use of fully homomorphic encryption in order to achieve privacy-preserving multi-party reconciliation on ordered sets. Assuming a sufficiently efficient encryption scheme, our solution performs much better than existing approaches in terms of communication overhead and number of homomorphic operations.