Search the digital library catalog

Abstract: V magistrskem delu obravnavamo razvoj in primerjavo različnih pristopov priporočilnih sistemov na področju knjižničnih sistemov. Najprej so predstavljeni teoretična izhodišča, uporabljene tehnologije in obdelava podatkov, ki so služili kot podlaga za implementacijo. V nadaljevanju so razviti in analizirani štirje pristopi: sodelovalno filtriranje, vsebinsko filtriranje ter mešani in kaskadni hibridni pristop. Eksperimentalni del temelji na vrednotenju z uporabo metrik HR@K, MRR@K in NDCG@K. Rezultati kažejo, da hibridni pristopi presegajo osnovne metode. Med njimi se je kot najbolj učinkovit izkazal kaskadni hibridni model, ki najbolje združuje prednosti obeh osnovnih tehnik in se prilagaja značilnostim knjižničnih podatkov.
Keywords: hibridni priporočilni sistem, knjižnični sistem, sodelovalno filtriranje, vsebinsko filtriranje, obdelava naravnega jezika
Published in DKUM: 17.10.2025; Views: 0; Downloads: 15
Full text (2,17 MB)

Abstract: V magistrskem delu predstavljamo implementacijo lastnega računalniškega igralca šaha po metodologiji AlphaZero, ki se igre nauči brez predhodnega znanja z igranjem proti samemu sebi. V prvem delu na kratko predstavimo zgodovino računalniškega igranja šaha ter izvedemo pregled osnovnih pristopov šahovskih pogonov. Drugi del zajema implementacijo, učenje in preverjanje delovanja računalniškega igralca. Računalniški igralec je dosegel 2144 točk ELO v igrah z igralnim časom ene minute in 1976 točk ELO v kategoriji s časom treh minut, kar ustreza nivoju močnega amaterskega igralca.
Keywords: šah, računalniški igralec, šahovski pogoni, iskalna drevesa, nevronske mreže
Published in DKUM: 17.10.2025; Views: 0; Downloads: 6
Full text (3,04 MB)

Abstract: We explore the sparsity and prunability of multi-layer perceptrons (MLPs) trained using the Forward-Forward (FF) algorithm, an alternative to backpropagation (BP) that replaces the backward pass with local, contrastive updates at each layer. We analyze the sparsity of the weight matrices during training using multiple metrics, and test the prunability of FF networks on the MNIST, FashionMNIST and CIFAR-10 datasets. We also propose FFLib—a novel, modular PyTorch-based library for developing, training and analyzing FF models along with a suite of FF-based architectures, including FFNN, FFNN+C and FFRNN. In addition to structural sparsity, we describe and apply a new method for visualizing the functional sparsity of neural activations across different architectures using the HSV color space. Moreover, we conduct a sensitivity analysis to assess the impact of hyperparameters on model performance and sparsity. Finally, we perform pruning experiments, showing that simple FF-based MLPs exhibit significantly greater robustness to one-shot neuron pruning than traditional BP-trained networks, and a possible 8-fold increase in compression ratios while maintaining comparable accuracy on the MNIST dataset.
Keywords: Forward-Forward, sparsity, pruning, model compression, machine learning, neural network
Published in DKUM: 20.08.2025; Views: 0; Downloads: 2
Full text (7,01 MB)

Abstract: Virtual pruning of simulated fruit tree models is a useful functionality provided by software tools for computer-aided horticultural education and research. It also enables algorithmic pruning optimization with respect to a set of quantitative objectives, which is important for analytical purposes and potential applications in automated pruning. However, the existing studies in pruning optimization focus on a single type of objective, such as light distribution within the crown. In this paper, we propose the use of heterogeneous objectives for discrete multi-objective optimization of simulated tree pruning. In particular, the average light intake, crown shape, and tree balance are used to observe the emergence of different pruning patterns in the non-dominated solution sets. We also propose the use of independent constraint objectives as a new mechanism to confine overfitting of solutions to individual pruning criteria. Finally, we perform the comparison of NSGA-II, SPEA2, and MOEA/D-EAM on this task. The results demonstrate that SPEA2 and MOEA/D-EAM, which use external solution archives, can produce better sets of non-dominated solutions than NSGA-II.
Keywords: multi-objective optimization, virtual tree pruning, heterogeneous objectives, constraint objectives, NSGA-II, SPEA2, EuMOEA/D-EAM
Published in DKUM: 19.06.2025; Views: 0; Downloads: 8
Full text (5,75 MB)
This document has many files! More...

Abstract: A new algorithm is presented for detecting the global reflection symmetry of geometric objects. The algorithm works for 2D and 3D objects which may be open or closed and may or may not contain holes. The algorithm accepts a point cloud obtained by sampling the object’s surface at the input. The points are inserted into a uniform grid and so-called boundary cells are identified. The centroid of the boundary cells is determined, and a testing symmetry axis/plane is set through it. In this way, the boundary cells are split into two parts and they are faced with the symmetry estimation function. If the function estimates the symmetric case, the boundary cells are further split until a given threshold is reached or a non-symmetric result is obtained. The new testing axis/plane is then derived and tested by rotation around the centroid. This paper introduces three techniques to accelerate the computation. Competitive results were obtained when the algorithm was compared against the state of the art.
Keywords: computer science, computational geometry, uniform subdivision, centroids
Published in DKUM: 01.04.2025; Views: 0; Downloads: 8
Full text (2,99 MB)
This document has many files! More...