Author name cluster

Bogdan Robu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

TAAS Journal 2026 Journal Article

Autonomic Resource Harvesting in HPC: Control Methods and Their Reusability

Quentin Guilloteau
Raphaël Bleuse
Sophie Cerf
Bogdan Robu
Rosa Pagano
Éric Rutten

High Performance Computing (HPC) systems are subject to dynamical variations occurring in, e.g., jobs execution duration, I/O quantity, network consumption. Adapting to these unpredictable variations requires using autonomic management in an online feedback loop. The introduction of control theory methods allows for the design of well-founded autonomic managers. Choosing the relevant approach is daunting due to the variety of existing controllers. The criteria are of different natures, involving performance and efficiency, but also required expertise in control theory, and reusability or portability between sub-systems. Therefore, there is a need for comparative studies to assist designers choices. We consider the problem of resource harvesting in HPC systems, where scheduling often leaves resources idle. Our approach controls—through a feedback loop—the injection of small jobs in order to maximize the resources’ usage. The control problem is to manage the tradeoff between harvesting and performance, in a reusable manner. We study how reusability relates to the adaptivity and robustness properties in control. We illustrate our approach with the classic Proportional-Integral-Derivative (PID) control, its upgrade as adaptive control, and Model-Free Control (MFC). We target CiGri, a system harvesting idle resources in a computing grid. We perform experimental evaluation and compare performance and reusability. Tradeoffs are found on different criteria: While adaptive control is largely portable, its design complexity is significant for non-experts; PID control has good nominal performance, yet its portability is limited; MFC requires few competences to be used, but cannot provide strong guarantees.

Details DOI

IJCAI Conference 2024 Conference Paper

A Conservative Approach for Few-Shot Transfer in Off-Dynamics Reinforcement Learning

Paul Daoudi
CHRISTOPHE PRIEUR
Bogdan Robu
Merwan Barlier
Ludovic Dos Santos

Off-dynamics Reinforcement Learning (ODRL) seeks to transfer a policy from a source environment to a target environment characterized by distinct yet similar dynamics. In this context, traditional RL agents depend excessively on the dynamics of the source environment, resulting in the discovery of policies that excel in this environment but fail to provide reasonable performance in the target one. In the few-shot framework, a limited number of transitions from the target environment are introduced to facilitate a more effective transfer. Addressing this challenge, we propose an innovative approach inspired by recent advancements in Imitation Learning and Conservative RL algorithms. This method introduces a penalty to regulate the trajectories generated by the source-trained policy. We evaluate our method across various environments representing diverse off-dynamics conditions, where access to the target environment is extremely limited. These experiments include high-dimensional systems relevant to real-world applications. Across most tested scenarios, our proposed method demonstrates performance improvements compared to existing baselines.

PDF Details DOI

EWRL Workshop 2024 Workshop Paper

A Conservative Approach for Few-Shot Transfer in Off-Dynamics Reinforcement Learning

Paul Daoudi
CHRISTOPHE PRIEUR
Bogdan Robu
Merwan Barlier
Ludovic Dos Santos

Off-dynamics Reinforcement Learning (ODRL) seeks to transfer a policy from a source environment to a target environment characterized by distinct yet similar dynamics. In this context, traditional RL agents depend excessively on the dynamics of the source environment, resulting in the discovery of policies that excel in this environment but fail to provide reasonable performance in the target one. In the few-shot framework, a limited number of transitions from the target environment are introduced to facilitate a more effective transfer. Addressing this challenge, we propose a new approach inspired by recent advancements in Imitation Learning and conservative RL algorithms. The proposed method introduces a penalty to regulate the trajectories generated by the source-trained policy. We evaluate our method across various environments representing diverse off-dynamics conditions, where it demonstrates performance improvements compared to existing baselines across most tested scenarios.

PDF

AAMAS Conference 2023 Conference Paper

Enhancing Reinforcement Learning Agents with Local Guides

Paul Daoudi
Bogdan Robu
CHRISTOPHE PRIEUR
Ludovic Dos Santos
Merwan Barlier

This paper addresses the problem of integrating local guide policies into a Reinforcement Learning agent. For this, we show how to adapt existing algorithms to this setting before introducing a novel algorithm based on a noisy policy-switching procedure. This approach builds on a proper Approximate Policy Evaluation (APE) scheme to provide a perturbation that carefully leads the local guides towards better actions. We evaluated our method on a set of classical Reinforcement Learning problems, including safetycritical systems where the agent cannot enter some areas at the risk of triggering catastrophic consequences. In all the proposed environments, our agent proved to be efficient at leveraging those policies to improve the performance of any APE-based Reinforcement Learning algorithm, especially in its first learning stages.

PDF

EAAI Journal 2023 Journal Article

Sparse dynamical features generation, application to Parkinson’s disease diagnosis

Houssem Meghnoudj
Bogdan Robu
Mazen Alamir

This study focuses on the diagnosis of Parkinson’s Disease (PD) based on electroencephalogram (EEG) signals. A novel approach inspired by the functioning of the brain is proposed, which uses the dynamics, frequency, and temporal content of EEGs to extract new discriminant features of the disease. The generated sparse dynamic features (SDFs) allow, through a transformation, to change the point of view on the data giving access to more informative features that are more faithful to the concept of EEG generation. Nevertheless, the method remains generic and can be applied to any signal but for this application it was evaluated on a publicly available dataset containing EEG signals recorded during a 3-oddball auditory task involving N = 50 subjects, of whom 25 suffer from PD. Given the adequate perspective on the data, it comes out that by using only two extracted features from the generated SDFs the healthy and unhealthy subjects are separated using a linear classifier. The classification yields an accuracy of 90. 0% ( p < 0. 03 ) using a single channel. By aggregating the information from three channels and making them vote, an accuracy of 94%, a sensitivity of 96% and a specificity of 92% is obtained. The evaluation was carried out using a nested Leave-One-Out cross-validation procedure, thus preventing data leakage problems and giving a less biased evaluation. Several tests were carried out to assess the validity and robustness of our approach.

Details DOI