All Years Seminars
2025-04-16 (15h) : Gradient-based optimization in CFD: from nuclear fusion to small modular fission reactors
At Euler building (room A.002)
Speaker :
Niels Horsten (IMMC,UCLouvain)
Abstract :
Gradient-based optimization is a powerful tool for advancing engineering applications. Its success relies on three key ingredients: accurate simulation models, efficient gradient calculations, and carefully chosen optimization methods. In this seminar, I will demonstrate how I developed fast, physics-based CFD models for nuclear fusion that are now being used for the design of future reactors. A particular focus will be on the challenges and solutions for obtaining gradients from Monte Carlo particle simulations, which are critical for modeling neutral-plasma interactions in fusion and neutron transport in fission systems. Lastly, I will discuss the application of these methods to the design of heat exchangers for lead-cooled small modular reactors (SMRs). The unique properties of liquid lead, particularly its low Prandtl number, introduce additional complexities for the optimization.
2025-04-15 (14h) : Open problems about the simplex method
At Euler building (room A.002)
Speaker :
Sophie Huiberts (LIMOS, Clermont Auvergne University)
Abstract :
The simplex method is a very efficient algorithm. In this talk we see a few of the state-of-the-art theories for explaining this observation. We will discuss what it takes for a mathematical model to explain an algorithm’s qualities, and whether existing theories meet this bar. Following this, we will question what the simplex method is and if the theoretician's simplex method is the same algorithm as the practitioner's simplex method. Along the way I will share some anecdotes about linear programming history.
2025-04-09 (15h) : Stochastic second-order optimization: global bounds, subspaces, and momentum
At Euler building (room A.002)
Speaker :
Doikov, Nikita
Abstract :
In this talk, we present stochastic second-order algorithms for solving general non-convex optimization problems. Using the cubic regularization, we prove global convergence rates for our methods. We will discuss two techniques that improve the properties of our algorithms in large-scale cases: stochastic subspaces, to deal with high-dimensional problems, and stochastic methods with momentum. The latter technique provably improves the variance of stochastic estimates and allows the method to converge for any noise level. This is in stark contrast to all existing stochastic second-order methods for non-convex problems, which typically require large batches.
2025-04-08 (14h) : Multi-product Supply Function Equilibria
At Euler building (room A.002)
Speaker :
Bert Willems (UCLouvain-LIDAM)
Abstract :
We solve for Nash equilibria in a procurement auction with multiple heterogeneous
divisible goods. There are (dis)economies of scope in production and goods
could be substitutes or complements for the procurer. Before demand is realized,
each firm offers a vector of supply functions where supply of a good depends on
the prices of all goods. This is related to the organization of the product-mix auction
and electricity markets with complex bids. We show that outcomes are not
influenced by bundling of the goods. For quadratic costs and linear demand, we
can use this property to transform the multi-product problem into an equivalent set
of separated markets, which can be analyzed independently. We show that Lerner
and pass-through tensors can be used to characterize mark-ups and welfare losses
in a multi-product market. Eigenvalues of the tensors are fundamental properties,
as they do not depend on bundling.
2025-03-31 (11h) : ICTEAM Colloquium series
At Maxwell building (room Shannon A.105)
Speaker :
Sepulchre, Rodolphe
Abstract : Regulation theory is grounded in the internal model principle, which states that exact regulation requires an exact internal model of the external signals to be regulated. How to reconcile this calibration principle with systems made of uncertain and variable components ? How do animals achieve regulation in changing and complex environments?
In this talk, Professor Sepulchre will propose that reliable regulation is possible in uncertain machines that regulate events rather than trajectories. He will highlight the role of excitability and synaptic coupling in a theory of event regulation
.
Seminars
INMA Contact Info
Mathematical Engineering (INMA)
L4.05.01
Avenue Georges Lemaître, 4
+32 10 47 80 36
secretaire-inma@uclouvain.be
Mon – Fri 9:00A.M. – 5:00P.M.
JOIN US