Organizers: Paul Irofti, Laurențiu Leuștean and Andrei Pătraşcu

The LOS seminar is the working seminar of the LOS research center.

All seminars, except where otherwise indicated, will be Tuesdays between 14:00 and 15.50. The seminars are held locally at Hall 214 (“Google”) of the Faculty of Mathematics and Computer Science, University of Bucharest, but can also be occasionally held remotely.

To receive announcements about the seminar, please send an email to los@fmi.unibuc.ro.

Tuesday, November 22, 2022

Paul Irofti (LOS)
Dictionary Learning with Uniform Sparse Representations for Anomaly Detection

Abstract: Many applications like audio and image processing show that sparse representations are a powerful and efficient signal modeling technique. Finding an optimal dictionary that generates at the same time the sparsest representations of data and the smallest approximation error is a hard problem approached by dictionary learning (DL). We study how DL performs in detecting abnormal samples in a dataset of signals. In this paper we use a particular DL formulation that seeks uniform sparse representations model to detect the underlying subspace of the majority of samples in a dataset, using a K-SVD-type algorithm. Numerical simulations show that one can efficiently use this resulted subspace to discriminate the anomalies over the regular data points.
This is joint work with Cristian Rusu and Andrei Pătrașcu.

Tuesday, October 18, 2022

Dongwon Lee (Pennsylvania State University)
Generative Language Model, Deepfake, and Fake News 2.0: Scenarios and Implications

Abstract: The recent explosive advancements in both generative language models in NLP and deepfake-enabling methods in Computer Vision have greatly helped trigger a new surge in AI research and introduced a myriad of novel AI applications. However, at the same time, these new AI technologies can be used by adversaries for malicious usages, opening a window of opportunity for fake news creators and state-sponsored hackers. In this talk, I will present three plausible scenarios where adversaries could exploit these cutting-edge AI techniques to their advantage, producing more sophisticated fake news by synthesizing realistic artifacts or evading detection of fake news from state-of-the-art detectors. I will conclude the talk by discussing the important implications of the new type of fake news (i.e., Fake News 2.0) and some future research directions.