ARTIFICIAL INTELLIGENCE Seminar

 

PROGRAM

---

Predavanja možete pratiti i online putem MITEAM stranice Seminara iz veštačke inteligencije:
https://miteam.mi.sanu.ac.rs/asset/DPP9i2jhvYzp8dmRe

---

Plan rada Seminara iz veštačke inteligencije za FEBRUAR 2026.

---

Sreda, 18.02.2026. u 19:00, Online
Nataša Krejić, Department of Mathematics and Informatics, Faculty of Science, University of Novi Sad
DISTRIBUTED INEXACT NEWTON METHOD WITH ADAPTIVE STEPSIZES
We consider two formulations for distributed optimization wherein _N_ nodes in a generic connected network solve a problem of common interest: distributed personalized optimization and consensus optimization. A new method termed DINAS (Distributed Inexact Newton method with Adaptive step size) is proposed. DINAS employs large adaptively computed step sizes, requires a reduced global parameters knowledge with respect to existing alternatives, and can operate without any local Hessian inverse calculations nor Hessian communications. When solving personalized distributed learning formulations, DINAS achieves quadratic convergence with respect to computational cost and linear convergence with respect to communication cost, the latter rate being independent of the local functions condition numbers or of the network topology. When solving consensus optimization problems, DINAS is shown to converge to the global solution. Extensive numerical experiments demonstrate significant improvements of DINAS over existing alternatives. As a result of independent interest, we provide for the first time convergence analysis of the Newton method with the adaptive Polyak's step size when the Newton direction is computed inexactly in a centralized environment.

---

Sreda, 25.02.2026. u 19:00, Online
Birojit Das, Department of Mathematics, National Institute of Technology (NIT) Agartala, India
CONVERGENCE AND MATRIX OPERATORS ON COMPLEX UNCERTAIN SEQUENCE SPACES
Uncertainty is an inherent feature of modern data-driven systems arising from noisy measurements, incomplete information, and stochastic environments. To rigorously analyze such situations, uncertainty theory provides a mathematical framework that extends classical probability-based modeling. In this talk, we develop a functional analytic foundation for handling sequences of complex uncertain variables through the study of infinite matrix transformations on uncertain sequence spaces.
We first introduce several notions of convergence for complex uncertain series—including convergence in mean, measure, distribution, almost surely, and uniformly almost surely—and establish their structural properties. Emphasis is placed on uniformly almost sure convergence, which proves particularly suitable for studying stability under linear transformations. We then characterize infinite matrix operators acting between uncertain sequence spaces and derive necessary and sufficient conditions for boundedness, linearity, and limit preservation. In this context, classical results such as the Silverman–Toeplitz and Kojima–Schur theorems are extended to the setting of complex uncertain sequences. Furthermore, summability methods and operator norms are investigated to ensure robustness of transformations.
Although the framework is theoretical, it provides essential mathematical guarantees for the reliability of data transformations under uncertainty. Such foundations are crucial for modern applications where algorithms must process noisy sensor signals, unreliable financial data, or incomplete medical information. By establishing stability and convergence criteria, this work contributes to the development of trustworthy and wellposed computational and AI systems built upon uncertain data.

---

Ovaj onlajn seminar nastao je kao nastavak sastanka "Serbian AI Meeting" i zamišljen je da na njemu istraživači iz Srbije i iz dijaspore, kao i istraživači sa univerzteta, naučnih instituta i iz prakse predstavljaju naučne teme i rezultate iz oblasti veštačke inteligencije.
Link za svako pojedinačno predavanje biće dostavljen dan pre održavanja predavanja.

---

---