3 papers
math.OC2025
Sequential test sampling for stochastic derivative-free optimization
Anjie Ding, Francesco Rinaldi, Luis Nunes Vicente
In many derivative-free optimization algorithms, a sufficient decrease condition decides whether to accept a trial step in each iteration. This condition typically requires that th…
math.OC2025
Direct-search methods in the year 2025: Theoretical guarantees and algorithmic paradigms
K. J. Dzahini, F. Rinaldi, C. W. Royer +1
Optimizing a function without using derivatives is a challenging paradigm, that precludes from using classical algorithms from nonlinear optimization, and may thus seem intractable…
cs.LG2025
Exploring the Potential of Bilevel Optimization for Calibrating Neural Networks
Gabriele Sanguin, Arjun Pakrashi, Marco Viola +1
Handling uncertainty is critical for ensuring reliable decision-making in intelligent systems. Modern neural networks are known to be poorly calibrated, resulting in predicted conf…