[Dottorcomp] REMINDER: METaL Seminars, Ambrogio Maria Bernardelli, 26/02/2026, 14:30-15:30, Aula Beltrami

[Siam Student Chapter]

Dear all,

This is a gentle reminder about the upcoming* METal Seminar** - Mathematics
& Engineering Talks and Lectures*, on *February 26th, *from *14:30 to
15:30 *at the Department of Mathematics in Aula Beltrami.

Here are the details of the Seminar*:*

   - *Speaker*: Ambrogio Maria Bernardelli - Researcher at University of
   Pavia.


   - *Title*:* Optimizing AI: MILPs for discrete Neural Networks.*


   - *Where and when*: February 26th, *14:30-15:30*, Aula Beltrami,
   Department of Mathematics, University of Pavia.


   - *Abstract: * Training neural networks is typically done using
   gradient-based methods, which often require large datasets, significant
   computational resources, and careful hyperparameter tuning. In this
   presentation, we explore an alternative approach based on Mixed-Integer
   Linear Programming (MILP) to train discrete neural networks exactly,
   particularly in low-data settings. The focus is on few-bit neural networks,
   including Binarized Neural Networks (BNNs), whose weights are restricted to
   +1 and −1, and Integer-Valued Neural Networks (INNs), whose weights lie
   within a limited integer range {−P, ..., P}. These models are especially
   attractive because of their lightweight architecture and their ability to
   run on low-power devices, where computations can be implemented using
   simple Boolean or integer operations. A new multiobjective ensemble method,
   called BeMi, is introduced. Instead of training a single network to
   distinguish all classes, the approach trains one network for each pair of
   classes and combines their predictions using a majority voting scheme. The
   training process simultaneously optimizes accuracy, robustness to small
   input perturbations, and sparsity, reducing the number of active weights in
   the network. Experimental results on the MNIST dataset show significant
   improvements over previous solver-based approaches. While earlier methods
   achieved an average accuracy of 51.1%, the proposed ensemble method reaches
   68.4% accuracy when trained with 10 images per class and 81.8% accuracy
   with 40 images per class. At the same time, it removes up to 75.3% of the
   network connections, producing simpler and more efficient models.


If you are not able to attend in presence, you can follow online using the
following link https://meet.google.com/wrc-ibhm-giy

We look forward to your participation!

Stay updated on SIAM Chapter activities by visiting our website
<https://sites.google.com/view/siam-unipv/home> and *follow us* on LinkedIn
<http://linkedin.com/company/siam-chapter-unipv> and Instagram
<https://www.instagram.com/siam.chapter.unipv/>!

Sincerely,

SIAM Chapter Organization Committee