News

SLG Seminar – Paul Gauthier Noé – 18/01/2024

10 January 2024

On 18 January from 12 am, we will host a talk from Dr. Paul Gauthier Noé on « Explaining probabilistic predictions … ». The presentation will be hosted on room S6.    More details will follow   Bio: Paul Gauthier Noe just received a PhD in Computer Science in Avignon Université under the supervision of Prof. Jean-François Bonastre and Dr. Driss Matrouf. He was working for the international JST-ANR VoicePersonae project and his main research interests are Speaker verification, Bayesian decision theory, Calibration of probabilities and Privacy in Speech.

PhD defense of Noé Cécillon – 18 January 2024

8 January 2024

Date: Thursday, January 18, 2023 at 14:00. Place: University of Avignon, Campus JH Fabre, Ada Lovelace amphitheater Jury : Title: Combining Graph and Text to Model Conversations: An Application to Online Abuse Detection. Abstract: Online abusive behaviors can have devastating consequences on individuals and communities. With the global expansion of internet and the social networks, anyone can be confronted with these behaviors. Over the past few years, laws and regulations have been established to regulate this kind of abuse but the responsibility ultimately lies with the platforms that host online communications. They are asked to monitor their users in order to prevent the proliferation of abusive content. Timely detection and moderation is a key factor to reduce the quantity and impact of abusive behaviors. However, due to the sheer quantity of online messages posted every day, platforms struggle to provide adequate resources. Since this implies high human and financial costs, companies have a keen interest in automating this process. Although it may seem a relatively simple task, it turns out to be quite complex. Indeed, malicious users have developed numerous techniques to bypass the standard automated methods. Allusions or implied meaning are other examples of strategies that automatic methods struggle Plus d'infos

SLG Seminar – Fenna Poletiek – 12/01/2024

8 January 2024

On 12 January from 12 am, we will host a virtual talk from Dr. Fenna Poletiek from Institute of Psychology at Leiden University on « Language learning in the lab ».   The presentation will be hosted on room S6.   Abstract: Language learning in the lab Language learning skills have been considered a defining feature of humanness. In this view language cannot be acquired by mere associative or statistical learning processes, only, like many other skills are learned by human and nonhuman primates during development. Indeed, the high (recursive) complexity of human grammars have been shown to make them impossible to learn by exposure to language exemplars only. Some research suggests, however, that at least some statistical learning is recruited in language acquisition (Perruchet & Pacton, 2006). And primates have been shown to mimic complex grammatical patterns after being trained on a sequence of stimulus responses (Rey et al., 2012). We performed series of studies with artificial languages in the lab, to investigate associative and statistical learning processes that support language learning. The results thus far suggest a fine tuned cooperation between three crucial features of the natural language learning process: first, learning proceeds ‘starting small’ with short simple sentences growing in complexity Plus d'infos

PhD defense of Julio Perez-Garcia – 18 December 2023

14 December 2023

Place: University of Avignon, Campus Hannah Arendt, Salle des ThèsesDate: Monday, December 18, 2023 at 14:00. Title: Contribution to security and privacy in the Blockchain-based Internet of Things: Robustness, Reliability, and Scalability. Abstract: The Internet of Things (IoT) is a diverse network of objects or ”things” typically interconnected via the Internet. Given the sensitivity of the information exchanged in IoT applications, it is essential to guarantee security and privacy. This problem is aggravated by the open nature of wireless communications, and the power and computing resource limitations of most IoT devices. At the same time, existing IoT security solutions are based on centralized architectures, which raises scalability issues and the single point of failure problem, making them susceptible to denial-of-service attacks and technical failures. Blockchain has emerged as an attractive solution to IoT security and centralization issues. Blockchains replicate a permanent, append-only record of all transactions occurring on a network across multiple devices, keeping them synchronized through a consensus protocol. Blockchain implementation may involve high computational and energy costs for devices. Consequently, solutions based on Fog/Edge computing have been considered in the integration with IoT. This approach shifts the higher computational load and higher energy consumption to the devices with higher Plus d'infos

SLG Meeting – St Germes Bengono Obiang – 21/12/2023

12 December 2023

The next SLG meeting will be held in room S1 on Thursday, December 21st, from 12:00 PM to 1:00 PM.    We will have the pleasure of hosting St Germes BENGONO OBIANG, a PhD student in speech processing, focusing on tone recognition in under-resourced languages. He is supervised by Norbert TSOPZE and Paulin MELATAGIA from the University of Yaoundé 1, as well as by Jean-François BONASTRE and Tania JIMENEZ from LIA.   Abstract: Many sub-Saharan African languages are categorized as tone languages and for the most part, they are classified as low resource languages due to the limited resources and tools available to process these languages. Identifying the tone associated with a syllable is therefore a key challenge for speech recognition in these languages. We propose models that automate the recognition of tones in continuous speech that can easily be incorporated into a speech recognition pipeline for these languages. We have investigated different neural architectures as well as several features extraction algorithms in speech (Filter banks, Leaf, Cestrogram, MFCC). In the context of low-resource languages, we also evaluated Wav2vec models for this task. In this work, we use a public speech recognition dataset on Yoruba. As for the results, using the Plus d'infos

PhD defense of Anais Chanclu – 11 December 2023

11 December 2023

Thesis defense of Anais Chanclu Date: Monday 11 December 2023 at 14:30  Location: Thesis room, Hannah Arendt campus. Title: Recognizing individuals by their voice: defining a scientific framework to ensure the reliability of voice comparison results in forensic contexts Jury: Abstract: In police investigations or criminal trials, voice recordings are often collected for comparison purposes with the voice of suspects. Typically, these recordings, referred to as ‘traces’, come from phone taps, emergency service calls, or voicemail messages. Recordings of suspects, known as ‘comparison pieces’, are usually obtained by law enforcement through voice sampling. Since the traces and comparison pieces were not recorded under the same conditions, and the recording conditions of the traces are often poorly known or entirely unknown, the variability between the recordings being compared cannot be quantified. Numerous factors come into play, including audio file characteristics, linguistic content, the recording environment, and the speaker(s). Voice comparison practices have evolved throughout history without conforming to a scientific framework. This has led to questioning the reliability of voice expertise (as in the Trayvon Martin case) and the use of fallacious practices (as in the Élodie Kulik case), potentially leading to judicial errors. Nowadays, the French Scientific Police (SNPS) and the Plus d'infos

Cornet Seminar – Andrea Fox – 08/12/2023

22 November 2023

In the context of team Cornet’s seminars, Andrea Fox (LIA) will present his research work on Safe Reinforcement Learning for Video Admission Control, on December 8, 2023, at 11:35 in the meeting room. Abstract: Mobile video cameras have become a pervasive commodity and represent an important candidate source to enhance video analytic applications. Yet, while available in large quantities, the limitations of the edge computing infrastructure require the careful selection of which video flows to process at any point in time to maximize the amount of information extracted by deployed applications. In this paper, we present an admission control scheme for mobile video streams originating from different areas and dispatched to multiple processing servers over an edge computing infrastructure. We introduce a model rooted in the theory of Constrained Markov Decision Processes (CMDPs) that captures the problem of ensuring adequate area coverage to applications, while accounting for constraints of edge servers and access network capacity. On top of this model, we develop two new policies based on specialized primal-dual constrained Reinforcement Learning methods that solve the optimal admission control problem. The first, called DR-CPO, adopts reward decomposition reinforcement learning. This technique effectively mitigates state-space explosion, achieves optimality, and significantly accelerates Plus d'infos

Cornet Seminar – Olivier Bilenne – 24/11/2023

22 November 2023

In the context of team Cornet’s seminars, Olivier Bilenne (LIA) will present his research work on Implementing fictitious play in partially observable stochastic games, on November 24, 2023, 11:35 in the meeting room. Abstract: Extensions of fictitious play to stochastic games have been recently examined in combination with reinforcement learning techniques inherent to Markov decision processes. We revisit this approach in the context of partially observable stochastic games. For this, we consider a two-player (finite-state) zero-sum stochastic game where one player (the attacker) has full visibility of the system, whereas the other player (the defender) has no access to the state of the opponent and must instead compose with public sources of information (in our setting: the actions played and their associated payoffs). We study a fictitious play dynamics where the players best response to the estimated empirical frequencies of action of their opponent. This sequence of play requires from the players to form beliefs on both their opponent’s strategy and on their own continuation payoff (modeled by a Q-function), based on the (full or partial) information that is available to them. The strategy estimation scheme, in particular, features a correction mechanism making up for delayed symptoms in the partially Plus d'infos

14th Conference on Decision and Game Theory for Security (GameSec-23)

11 November 2023

The 14th Conference on Decision and Game Theory for Security (GameSec-23) will take place from October 18-20, 2023 in Avignon, France. With the rapid development of information, automation, and communication technology, the security of these emerging systems is more important now than ever. GameSec 2023 focuses on the protection of heterogeneous, large-scale, and dynamic cyber-physical systems as well as managing security risks faced by critical infrastructures through rigorous and practically relevant analytical methods. GameSec 2023 invites novel, high-quality theoretical and empirical contributions, which leverage decision theory and game theory to address security problems and related problems such as privacy, trust, or bias in emerging systems. The goal of the conference is to bring together academic, government, and industrial researchers in an effort to identify and discuss the major challenges and recent results that highlight the interdisciplinary connections between game theory, control, distributed optimization, adversarial reasoning, machine learning, mechanism design, behavioral analysis, risk assessments, and security, reputation, trust and privacy problems. Website: www.gamesec-conf.org

PhD defense of Thibault Cordier – 13 October 2023

13 October 2023

Date: Friday, the 13th of October at 9 am, Place: room “salle des thèses” at l’Université d’Avignon, Campus Hannah Arendt (centre-ville). Title: « Hierarchical Imitation and Reinforcement Learning for Multi-Domain Task-Oriented Dialogue Systems ». The defense can be followed through the live link below: https://v-au.univ-avignon.fr/live Abstract: In this Ph.D thesis, we study task-oriented dialogue systems that are systems designed to assist users in completing specific tasks, such as booking a flight or ordering food. They typically rely on reinforcement learning paradigm to model the dialogue that allows the system to reason about the user’s goals and preferences, and to select actions that will lead to the desired outcome. Our focus is specifically on learning from a limited number of interactions that is crucial due to the scarcity and costliness of human interactions. Standard reinforcement learning algorithms typically require a large amount of interaction data to achieve good performance. To address this challenge, we aim to make dialogue systems more sample-efficient in their training. We draw from two main ideas: imitation and hierarchy. Our first contribution explores the integration of imitation with reinforcement learning. We investigate how to effectively use expert demonstrations to extrapolate knowledge with minimal generalisation effort. Our second contribution focuses on Plus d'infos

1 2 3 4 … 8

Laboratoire Informatique d'Avignon — Avignon Université