Embedded Files
Speaker: Terry Biths Mongkuo Yanoh, YIEL RH.
Title: When AI meets Industry 4.0
Abstract: With the ever increasing solicitations on the precision and flexibility of robotic arms in the manufacturing industry, we investigate how the combination of robust model predictive control and physics-informed neural networks can improve the control of a two-link robot arm. This approach enables the robot to follow the desired path accurately despite uncertainties and disturbances bounded by an ellipsoid.
Date & Time: [06th April 2026 | 4 PM (WAT)]
Venue: Online (Google Meet Link: meet.google.com/pjq-xson-qow )
Speaker: Michel Roland BIDIAS A KESSENG, Kes-Africa.
Title: Robust Nonlinear Control of a Quadrotor UAV Using a Hybrid Backstepping–Sliding Mode Approach.
Abstract: Quadrotor unmanned aerial vehicles (UAVs) exhibit highly nonlinear, coupled, and under-actuated dynamics, making robust control design particularly challenging in the presence of disturbances and model uncertainties. In this work, I develop a six-degree-of-freedom dynamic model of a quadrotor that explicitly incorporates motor dynamics. I then propose a hybrid nonlinear control strategy combining backstepping for position tracking and sliding mode control for attitude stabilization within a two-loop architecture. Numerical simulations under nominal conditions, external disturbances, and parametric uncertainties demonstrate accurate trajectory tracking, strong disturbance rejection, and limited chattering effects. The results highlight the effectiveness of hybrid nonlinear control strategies for improving the robustness and reliability of quadrotor UAV systems.
Date & Time: [09th March 2026 | 5 PM (WAT)]
Venue: Online (Google Meet Link: meet.google.com/pjq-xson-qow )
Speaker: Dr. Nourridine Siewe, Rochester Institute of Technology, United States of America.
Title: Modelling treatment of osteoarthritis with standard therapy and senolytic drugs.
Abstract: Osteoarthritis (OA), the most common form of joint disease, involves the progressive degradation of articular cartilage and is a major cause of chronic disability in aging populations. Since OA is associated with severe deficiency of collagen type II, clinical trials considered treatment of OA by injection with undenatured collagen type II (UC-II). Recent studies consider also injection of senolytic drugs, like fisetin, that eliminates senescent chondrocytes in aging patients, to reduce the negative effect of these senescent cells on cartilage structure. In this paper we develop a mathematical model of OA for men and, separately, for women, and use the model to assess the efficacy of treatment by UC-II and by fisetin, alone or in combination. Our computations show the benefits of starting treatment early. They also show that although the effect of treatment by fisetin on slowing the progression of OA is much smaller compared to UC-II treatment, its effect in combination with UC-II is significantly increased.
Date & Time: [23rd February 2026 | 5 PM (WAT)]
Venue: Online (Google Meet Link: meet.google.com/pjq-xson-qow )
Speaker: Dr. Gilles MADI WAMBA, AXA, France.
Title: From Calculus 101 to Foundation Models: How Gradient Descent Took Over the World.
Abstract: Gradient descent is often introduced as a simple optimization trick in early university math courses: take a derivative, follow the slope, repeat. Nothing fancy. And yet, this very idea sits at the core of modern artificial intelligence—powering deep neural networks with billions of parameters, large language models, computer vision systems, and today’s so-called foundation models. In this talk, I will revisit gradient descent from its most elementary mathematical formulation and follow its evolution into a central tool of modern machine learning. We will see how basic concepts from calculus, linear algebra, and optimization quietly scale up to train models that recognize faces, understand language, and make decisions in complex real-world systems. Along the way, I will connect classical mathematical intuition with practical ML engineering realities: loss landscapes, stochasticity, constraints, regularization, and why “just minimizing a function” turns out to be both incredibly powerful and surprisingly subtle. The goal of this talk is not to sell AI magic, but to show how solid bachelor-level mathematics—used well—still underpins today’s most advanced AI systems.
Date & Time: [9th February 2026 | 5 PM (WAT)]
Speaker: Dr. Patrice Ndambomve, University of Buea, Cameroon.
Title: Controllability of Dynamical Systems: From the Finite to the Infinite Dimensional Settings.
Abstract: A Control System is a dynamical system on which one can act by the use of suitable control parameters. While studying a control system, one of the most common problems that appear is the controllability problem, which consists in checking the possibility of steering the control system from an initial state (initial condition) to a desired terminal one (boundary condition), by an appropriate choice of the control u. The controllability problem has two distinguished notions, which are the exact and the approximate controllability problems. While in finite dimension these two notions coincide, it is not the case in infinite dimensions where the approximate controllability which is very often completely adequate in applications, is much weaker than the exact controllability which is stronger. In this talk, we explore these notions from linear systems to nonlinear differential and integrodifferential control systems, using techniques from fixed-point theory, semigroup theory, and resolvent operator for integral equations.
Date & Time: [26th January 2026 | 5 PM (WAT)]
Venue: Online (Google Meet Link: meet.google.com/pjq-xson-qow )
Speaker:, Dr. Dena Firoozi, University of Toronto.
Title: Mean Field Games in Infinite Dimensions.
Abstract: Originally developed in finite-dimensional spaces, mean field games (MFGs) have become pivotal in addressing large-scale problems involving numerous interacting agents, and have found extensive applications in economics and finance. However, there are scenarios where Euclidean spaces do not adequately capture the essence of a problem such as non-Markovian systems. A clear and intuitive example is systems involving time delays. This talk presents a comprehensive study of linear-quadratic (LQ) MFGs in Hilbert spaces, generalizing the classic LQ MFG theory to scenarios involving N agents with dynamics governed by infinite-dimensional stochastic equations. We first study the well-posedness of a system of N coupled semilinear infinite-dimensional stochastic evolution equations establishing the foundation of MFGs in Hilbert spaces. We then specialize to N-player LQ games described above and study the asymptotic behavior as the number of agents, N, approaches infinity. We develop an infinite-dimensional variant of the Nash Certainty Equivalence principle and characterize a unique Nash equilibrium for the limiting MFG. Finally, we study the connections between the N-player game and the limiting MFG, establishing the ϵ-Nash property for the resulting limiting best-response strategy. Finally, if time permits, we present results that incorporate an infinite-dimensional common noise. In this setting, the mean-field consistency condition is characterized by a system of forward–backward stochastic evolution equations, whereas in the absence of common noise, it reduces to a system of forward–backward deterministic evolution equations in Hilbert spaces. We establish the solvability of these equations and prove the ϵ‑Nash property of the equilibrium strategy within this extended framework.
Date & Time: [12th January 2026 | 5 PM (WAT)]
Venue: Online (Google Meet Link: meet.google.com/pjq-xson-qow )
Speaker:, Dr. Yaroslav Salii, Kinaxis Inc.
Title: From Academia to Industry: Journey of an Applied Mathematician.
Abstract: The talk will give a subjective mathematician's perspective on product development in software, and a no less subjective software engineer's perspective on research in mathematical optimization, with a focus on challenges specific to these worlds, and the lessons and skills transferable between them.
Date & Time: [17th November 2025 | 5 PM (WAT)]
Speaker:, Dr. Gökçe Dayanıklı, University of Illinois Urbana-Champaign.
Title: Utilizing deep learning and game theory to find optimal policies for a large number of agents.
Abstract: In this talk, we will discuss how we can utilize deep learning to solve complex (dynamic and stochastic) game theoretical problems where there are many agents (such as banks, companies or people) interacting. We will first look at a stochastic optimal control problem for one agent and explain how we can use deep learning to solve this problem. Later, we will move on to the multi-agent setup, and we will discuss and compare two equilibrium notions in game theory: Nash equilibrium and Stackelberg equilibrium. After explaining how a Nash equilibrium can be approximated for dynamic and stochastic games with a large number of agents through mean field games, we will introduce the Stackelberg mean field games between a principal (i.e., regulator) and many agents. Stackelberg mean field game models can be used to find incentives or optimal policies for a large group of noncooperative agents with a motivation to model different real life problems such as regulating the systemic risk in banking sector or mitigating an epidemic. We will discuss how (intrinsically bi-level) Stackelberg mean field game model can be rewritten to propose a single-level deep learning method to solve this complex problem and conclude with some examples.
Date & Time: [1th December 2025 | 5 PM (WAT)]
Speaker:, Dr. Nathalie Ayi, Sorbonne University.
Title: Large Population Limit for Interacting Particle Systems on Random and Deterministic Weighted Graphs.
Abstract: When studying interacting particle systems, two distinct categories emerge: indistinguishable systems, where particle identity does not influence system dynamics, and non-exchangeable systems, where particle identity plays a significant role. One way to conceptualize these second systems is to see them as particle systems on weighted graphs. In this talk, we focus on the latter category. Recent developments in graph theory have raised renewed interest in understanding large population limits in these systems. Two main approaches have emerged: graph limits and mean-field limits. While mean-field limits were traditionally introduced for indistinguishable particles, they have been extended to the case of non-exchangeable particles recently. In this presentation, we introduce several models, mainly from the field of opinion dynamics, for which rigorous convergence results as N tends to infinity have been obtained. We also clarify the connection between the graph limit approach and the mean-field limit one. The works discussed draw from several papers, some co-authored with Nastassia Pouradier Duteil and David Poyato.
Date & Time: [15th December 2025 | 5 PM (WAT)]
Page updated
Report abuse