2 code implementations • 5 Apr 2024 • Paul Irofti, Iulian-Andrei Hîji, Andrei Pătraşcu, Nicolae Cleju
We introduce a new anomaly detection model that unifies the OC-SVM and DL residual functions into a single composite objective, subsequently solved through K-SVD-type iterative algorithms.
1 code implementation • 5 Mar 2024 • Andrei Pătraşcu, Cristian Rusu, Paul Irofti
Sparsifying transforms became in the last decades widely known tools for finding structured sparse representations of signals in certain transform domains.
1 code implementation • 27 Nov 2023 • Luis Romero-Ben, Paul Irofti, Florin Stoican, Vicenç Puig
In this paper, we present a nodal hydraulic head estimation methodology for water distribution networks (WDN) based on an Unscented Kalman Filter (UKF) scheme with application to leak localization.
1 code implementation • 21 Apr 2023 • Paul Irofti, Luis Romero-Ben, Florin Stoican, Vicenç Puig
This article presents a leak localization methodology based on state estimation and learning.
no code implementations • 14 May 2022 • Paul Irofti, Andrei Pătraşcu, Andrei Iulian Hîji
Cyberthreats are a permanent concern in our modern technological world.
1 code implementation • 11 Jan 2022 • Paul Irofti, Cristian Rusu, Andrei Pătraşcu
In this paper we use a particular DL formulation that seeks uniform sparse representations model to detect the underlying subspace of the majority of samples in a dataset, using a K-SVD-type algorithm.
no code implementations • 12 Oct 2021 • Paul Irofti, Luis Romero-Ben, Florin Stoican, Vicenç Puig
In this paper, we propose a data-driven leak localization method for water distribution networks (WDNs) which combines two complementary approaches: graph-based interpolation and dictionary classification.
1 code implementation • 10 Aug 2021 • Andrei Patrascu, Paul Irofti
Several decades ago the Proximal Point Algorithm (PPA) started to gain a long-lasting attraction for both abstract operator theory and numerical optimization communities.
1 code implementation • 7 Jul 2020 • Cristian Rusu, Paul Irofti
Separable, or Kronecker product, dictionaries provide natural decompositions for 2D signals, such as images.
no code implementations • 30 Mar 2020 • Andrei Patrascu, Ciprian Paduraru, Paul Irofti
Stochastic optimization lies at the core of most statistical learning models.
1 code implementation • 18 Mar 2020 • Paul Irofti, Florin Stoican, Vicenç Puig
Fault detection and isolation in water distribution networks is an active topic due to its model's mathematical complexity and increased data availability through sensor placement.
no code implementations • 29 Feb 2020 • Paul Irofti, Andra Băltoiu
We investigate the possibilities of employing dictionary learning to address the requirements of most anomaly detection applications, such as absence of supervision, online formulations, low false positive rates.
1 code implementation • 4 Dec 2019 • Andrei Patrascu, Paul Irofti
In the large-scale or noisy contexts, when only stochastic information on the smooth part of the objective function is available, the extension of proximal gradient schemes to stochastic oracles is based on proximal tractability of the nonsmooth component and it has been deeply analyzed in the literature.
no code implementations • 24 Oct 2019 • Andra Baltoiu, Andrei Patrascu, Paul Irofti
Anomaly detection in networks often boils down to identifying an underlying graph structure on which the abnormal occurrence rests on.
no code implementations • 24 Oct 2019 • Paul Irofti, Andrei Patrascu, Andra Baltoiu
In general, anomaly detection is the problem of distinguishing between normal data samples with well defined patterns or signatures and those that do not conform to the expected profiles.
no code implementations • 16 Sep 2015 • Paul Irofti, Bogdan Dumitrescu
Dictionary learning for sparse representations is traditionally approached with sequential atom updates, in which an optimized atom is used immediately for the optimization of the next atoms.
no code implementations • 16 Dec 2014 • Paul Irofti
Dictionary training for sparse representations involves dealing with large chunks of data and complex algorithms that determine time consuming implementations.