The AIDA paper introduces an efficient and secure distributed K-Means algorithm The AIDA project consortium has published a paper at the 19th Symposium on Intelligent Data Analysis (IDA 2021), which took place online, between April 26 and 28. The article is entitled “Efficient Privacy Preserving Distributed K-Means for Non-IID Data” and it introduces an efficient and secure distributed K-Means algorithm, that is robust to non-IID data. The base idea of our proposal consists in each client computing the K-Means algorithm locally, with a variable number of clusters. The server will use the resultant centroids to apply the K-Means algorithm again, discovering […]
AIDA researchers presented a paper at PaPoC Workshop 2021 Researchers from the AIDA project presented a new paper at the 8th Workshop on Principles and Practice of Consistency for Distributed Data, which took place online, on April 26, together with EuroSys 2021. The article is entitled “Totally-Ordered Prefix Parallel Snapshot Isolation” and it provides transactional SQL semantics (i.e., NewSQL) in novel database systems, facilitating the migration of applications. TOPSI is a new algorithm for transaction isolation that avoids waiting for remote updates or using a stale snapshot. This is particularly useful in distributed cloud/edge database systems where wide-area communication or transient […]
The AIDA paper contributes to the trustworthiness of deep learning models The AIDA project consortium has published a paper at the 19th Symposium on Intelligent Data Analysis (IDA 2021), which took place online, between April 26 and 28. The article is entitled “Partially Monotonic Learning for Neural Networks” and it shows that it is possible to encode monotonic relations between variables in complex neural networks without significantly losing model accuracy. For example, when increasing the price of some goods or services, while all other variables remaining equal, real world knowledge indicates that sales will drop (or at least not increase). […]
AIDA researchers presented a paper at ACM SAC 2021 The AIDA project consortium has published a paper at the 2021 ACM/SIGAPP Symposium on Applied Computing (ACM SAC 2021), which took place online, between March 22 and 26. The article is entitled “Analysis of VM Migration Scheduling as Moving Target Defense against insider attacks”; the paper aims to study the security benefit of shuffling the attacker position in the environment through Moving Target Defense based on Virtual Machine migration (VM-MTD). The results show a significant reduction in the attack success probability when applying VM-MTD. It also presents a comparison between different […]
AIDA interview with COMPETE2020 AIDA’s research manager, Carlos Martins (Mobileum), was interviewed by COMPETE 2020 about the mission and objectives of the project. The interview is in Portuguese and available here.
CMU Team received awards and distinctions Over the past weeks, the partners from the Carnegie Mellon University received scientific awards in the fields related to AIDA. Fredkin Professorship in Artificial Intelligence Christos Faloutsos, in charge of leading the CMU collaboration with the AIDA project, and professor in the Computer Science Department of CMU, received the Fredkin Professorship in Artificial Intelligence during an online event on Thursday, Oct. 22. His interests include large-scale data mining with an emphasis on graphs and time sequences, anomaly detection, tensors, and fractals. An Association for Computing Machinery (ACM) fellow and an Amazon Scholar, […]
AIDA partners organized an internal workshop The partners from the University of Coimbra and Carnegie Mellon University organized an internal workshop on October 28 and, entitled “Distribution and Orchestration and Secure Communications”. This initiative focused on presenting the most recent research developed by both institutions, also allowing the members of both teams to discuss possible collaboration activities within the scope of the AIDA project.
AIDA has started The kickoff meeting of the AIDA project took place on May 5 2020 via a virtual platform, with approximately 20 participants from four academia and industry partners. The main objective of the $2M project is improving a platform used by Mobileum for integral risk management in companies. This platform ensures revenue, corporate conditions and fraud control for companies. “Managing risks and assuring business continuity, in an interconnected online digital world, where massive amounts of data are exchanged in real time, demands an equally disruptive approach where artificial intelligence, 5G and edge computing, in hybrid cloud architectures are […]