Master's program at the Faculty of Mathematics and Computer Science, University of Bucharest
Speaker: Andra Băltoiu (University of Bucharest) Abstract: In a previous seminar, we introduced Dictionary Learning (DL), a machine learning method capable of handling the requirements of IoT-related tasks, motivated by its reduced computational complexity, theoretical guarantees and its applicability to continuous retraining contexts. We now discuss the task of training different machine learning and DL …
Speaker: Ionuţ Ţuţu (Royal Holloway, University of London) Abstract: We propose a hybrid-dynamic first-order logic as a formal foundation for specifying and reasoning about reconfigurable systems. As the name suggests, the formalism we develop extends (many-sorted) first-order logic with features that are common to hybrid and to dynamic logics. This provides certain key advantages for …
Speaker: Georgel Gheorghe este absolvent al facultatii de Electronica, Telecomunicatii si Tehnologia a informatiei. Este membru ISACA si detine certificari precum CISA (Certified Information Systems Auditor), CGEIT (Certified in the Governance of Enterprise IT) si CRISC (Certified in Risk and Information Systems Control). Are o experienta de peste 11 ani in audit si consultanta in …
Speaker: Alexandru Dragomir (University of Bucharest) Abstract: Dynamic epistemic logics are useful in reasoning about knowledge and certain acts of learning (epistemic actions). However, not all epistemic actions are allowed to be executed in an initial epistemic model, and this is where the concept of a protocol comes in: a protocol stipulates what epistemic actions …
Speaker: Radu Velea (BitDefender) Abstract: Static code analysis methods have the advantage of providing deterministic and reliable results. Malware has evolved beyond the point where simple pattern matching algorithms or signatures can provide adequate levels of protection. To respond to new threats we have to look at other hidden aspects such as execution behavior and …
Speaker: Andrei Pătrașcu (University of Bucharest) Abstract: Recent empirical results confirm that one-class (OC) classification methods remain among the most important learning strategies for anomaly detection. In this seminar, we will technically describe in detail multiple basic OC schemes such as OC-SVM and SVDD and their deep variants, in order to identify room of improvements …
Speaker: Andrei Pătrașcu (University of Bucharest) Abstract: We continue our adventure by investigating existing results with Robust Principal Component Analysis (RPCA) and its adaptation to existing deep neural networks. Required reading: ZHOU, Chong; PAFFENROTH, Randy C. Anomaly detection with robust deep autoencoders. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and …
Speaker: Andra Băltoiu (University of Bucharest) Abstract: We continue our investigation on the task of detecting outliers in networks, by looking at the concept of signal variation on a graph. Required reading: A. Sandryhaila and J. M. F. Moura, "Classification via regularization on graphs," 2013 IEEE Global Conference on Signal and Information Processing, Austin, TX, …
Speaker: Paul Irofti (University of Bucharest) Abstract: We continue our investigation on the task of detecting outliers in networks when dealing with big-data and investigate existing online and distributed solutions. Required reading: Miao, Xuedan, et al. "Distributed online one-class support vector machine for anomaly detection over networks." IEEE transactions on cybernetics 49.4 (2018): 1475-1488. Liu, …
Speaker: Andrei Pătrașcu (University of Bucharest) Abstract: We continue our adventure by investigating existing results using Gaussian Mixture Models (GMM) for anomaly detection and their adaptation to existing deep neural networks. Required reading: Zong, Bo, et al. "Deep autoencoding gaussian mixture model for unsupervised anomaly detection." (2018). Chapter 11 from Deisenroth, Marc Peter, A. Aldo …