A5102943642- Reinforcement Learning in Robotics
- Bayesian Modeling and Causal Inference
- Formal Methods in Verification
- Multi-Agent Systems and Negotiation
- Machine Learning and Algorithms
- Robot Manipulation and Learning
- Human-Automation Interaction and Safety
- Auction Theory and Applications
- Context-Aware Activity Recognition Systems
- Non-Destructive Testing Techniques
- Logic, Reasoning, and Knowledge
- Welding Techniques and Residual Stresses
- AI-based Problem Solving and Planning
- Video Surveillance and Tracking Methods
- Corporate Governance and Law
- Human Pose and Action Recognition
- Reservoir Engineering and Simulation Methods
- Advanced Control Systems Optimization
- Seismic Imaging and Inversion Techniques
- Simulation Techniques and Applications
- Data Stream Mining Techniques
- Artificial Intelligence in Games
- Social Robot Interaction and HRI
- Interactive and Immersive Displays
- Social Sciences and Governance
Laboratoire Lorrain de Recherche en Informatique et ses Applications
2002-2024
Centre National de la Recherche Scientifique
2022-2024
Université de Lorraine
2010-2024
Beicip Franlab (France)
2016-2024
Iqra University
2024
Institut national de recherche en informatique et en automatique
2012-2023
Université de Bourgogne
1999-2017
Instituts Universitaires de Technologie
2013-2017
Siemens (Germany)
2016
Université Laval
2008-2010
The difficulties encountered in sequential decision-making problems under uncertainty are often linked to the large size of state space. Exploiting structure problem, for example by employing a factored representation, is usually an efficient approach but, case partially observable Markov decision processes, fact that some variables may be visible has not been sufficiently appreciated. In this article, we present complementary analysis and discussion about MOMDPs, formalism exploits space...
Model-based Bayesian Reinforcement Learning (BRL) allows a found formalization of the problem acting optimally while facing an unknown environment, i.e., avoiding exploration-exploitation dilemma. However, algorithms explicitly addressing BRL suffer from such combinatorial explosion that large body work relies on heuristic algorithms. This paper introduces BOLT, simple and (almost) deterministic algorithm for which is optimistic about transition function. We analyze BOLT's sample complexity,...
In human-robot collaboration, the objectives of human are often unknown to robot. Moreover, even assuming a known objective, behavior is also uncertain. order plan robust robot behavior, key preliminary question then: How derive realistic behaviors given objective? A major issue that such should itself account for otherwise collaboration cannot happen. this paper, we rely on Markov decision models, representing uncertainty over objective as probability distribution finite set functions...
Abstract Hydrocarbon exploration along the Arabian Peninsula is almost celebrating a century of successes. Major structures were drilled and hundreds billions barrels consequently discovered still producing at increasing rates. Remarkable multi-scale multi-disciplinary dataset (e.g., 2D, 3D seismic data, core well log data imagery…) have been acquired in past decades allowing geoscientists to better assess diverse onshore offshore Petroleum Systems’ potential. Many challenges linked new...
Work-related musculoskeletal disorders (WMSDs) are very common. Repetitive motion, which is often present in industrial work, one of the main physical causes WMSDs. It uses same set human joints repeatedly, leads to localized joint fatigue. In this we a framework plan policy collaborative robot that reduces fatigue long term, highly repetitive co-manipulation tasks, while taking into account uncertainty postural reaction motion and partial observability state. We model problem using...
The design of an eddy current imaging array probe dedicated to the inspection aeronautical fastener holes is presented.The aims at enhancing characterisation defects such as surface fatigue cracks well simplifying procedures (limited displacements).In this paper, a general featuring separate inducing and sensing functions proposed studied thanks 3D finite element computations.Then, first experimental validation structure given.Finally alternative pickup sensor technologies are considered.A...
A simplified electromagnetic modeling is proposed for the eddy current imaging of surface breaking defects. The model assumes that interactions between currents and a defect are equivalent to sources placed inside defect. use distributed point source method thus suitable implementing model. An example provided considering case aeronautical fastener holes. results given by confronted with experimental data obtained an probe as well finite elements simulations good agreement shown.
Reactive problem solving is a way to propose systems composed of simple interacting agents that collectively solve problems outside the scope individual perceptions. In this domain, natural social are sources inspiration for mechanisms.This article presents an approach region detection inspired by spiders. Based on behavioral model determined simulation collective weaving, we describe how transposed it obtain in gray level images and first assessment approach.
In this article, we discuss how to solve information-gathering problems expressed as rho-POMDPs, an extension of Partially Observable Markov Decision Processes (POMDPs) whose reward rho depends on the belief state. Point-based approaches used for solving POMDPs have been extended rho-POMDPs MDPs when its is convex in B or it Lipschitz-continuous. present paper, build POMCP algorithm propose a Monte Carlo Tree Search aiming efficient on-line planner which can be any function. Adaptations are...
In sequential decision making, heuristic search algorithms allow exploiting both the initial situation and an admissible to efficiently for optimal solution, often planning purposes. Such exist problems with uncertain dynamics, partial observability, multiple criteria, or collaborating agents. this article, we look at two-player zero-sum stochastic games (zsSGs) a discounted criterion, in view propose solution tailored fully observable case, while solutions have been proposed particular,...
This article presents a new formalism Interac-DEC-MDP whose aim is to introduce the concept of interaction in Decentralized Markov Decision Process and which has been inspired by biology. The this formalism, Interac-DEC-MDP, describe represent interactions among agents. outcome decided collectively two agents charge distribution local rewards. We have modeled biological experiment within formalism. A simple learning algorithm applied on generates more efficient collective behavior than...
This study endeavours to delve into the repercussions of organizational learning on knowledge performance educators within Catholic schools Karachi, Pakistan. The adopted a cross-sectional survey design, enlisting participation 282 teachers, situated secondary under jurisdiction Board Education in Karachi. data collection tool employed this was Dimensions Learning School Questionnaire (DLSQ). To analyse collected data, research harnessed statistical prowess Statistical Package Social...