Digitalni repozitorij raziskovalnih organizacij Slovenije

Iskanje po repozitoriju
A+ | A- | Pomoč | SLO | ENG

Iskalni niz: išči po
išči po
išči po
išči po

Možnosti:
  Ponastavi


Iskalni niz: "polno besedilo" AND "organizacija" (Institut Jožef Stefan) .

1 - 10 / 155
Na začetekNa prejšnjo stran12345678910Na naslednjo stranNa konec
1.
A cross-benchmark examination of feature-based algorithm selector generalization in single-objective numerical optimization
Gjorgjina Cenikj, Gašper Petelin, Tome Eftimov, 2024, izvirni znanstveni članek

Povzetek: The task of selecting the best optimization algorithm for a particular problem is known as algorithm selection (AS). This involves training a model using landscape characteristics to predict algorithm performance, but a key challenge remains: making AS models generalize effectively to new, untrained benchmark suites. This study assesses AS models’ generalizability in single-objective numerical optimization across diverse benchmark suites. Using Exploratory Landscape Analysis (ELA) and transformer-based (TransOpt) features, the research investigates their individual and combined effectiveness in AS across four benchmarks: BBOB, AFFINE, RANDOM, and ZIGZAG. AS models perform differently based on benchmark suite similarities in algorithm performance distributions and single-best solvers. When suites align, these models underperform against a baseline predicting mean algorithm performance; yet, they outperform the baseline when suites differ in performance distributions and solvers. The AS models trained using the ELA landscape features are better than the models trained using the TransOpt features on the BBOB and AFFINE benchmark suites, while the opposite is true for the RANDOM benchmark suite. Ultimately, the study reveals challenges in accurately capturing algorithm performance through problem landscape features (ELA or TransOpt), impacting AS model applicability.
Ključne besede: algorithm selection, multi-target regression, generalization, benchmarking
Objavljeno v DiRROS: 21.05.2024; Ogledov: 78; Prenosov: 222
.pdf Celotno besedilo (2,49 MB)
Gradivo ima več datotek! Več...

2.
Exploiting image quality measure for automatic trajectory generation in robot-aided visual quality inspection
Atae Jafari-Tabrizi, Dieter P. Gruber, Andrej Gams, 2024, izvirni znanstveni članek

Povzetek: Currently, the standard method of programming industrial robots is to perform it manually, which is cumbersome and time-consuming. Thus, it can be a burden for the flexibility of inspection systems when a new component with a different design needs to be inspected. Therefore, developing a way to automate the task of generating a robotic trajectory offers a substantial improvement in the field of automated manufacturing and quality inspection. This paper proposes and evaluates a methodology for automatizing the process of scanning a 3D surface for the purpose of quality inspection using only visual feedback. The paper is divided into three sub-tasks in the same general setting: (1) autonomously finding the optimal distance of the camera on the robot’s end-effector from the surface, (2) autonomously generating a trajectory to scan an unknown surface, and (3) autonomous localization and scan of a surface with a known shape, but with an unknown position. The novelty of this work lies in the application that only uses visual feedback, through the image focus measure, for determination and optimization of the motion. This reduces the complexity and the cost of such a setup. The methods developed have been tested in simulation and in real-world experiments and it was possible to obtain a precision in the optimal pose of the robot under 1 mm in translational, and 0.1° in angular directions. It took less than 50 iterations to generate a trajectory for scanning an unknown free-form surface. Finally, with less than 30 iterations during the experiments it was possible to localize the position of the surface. Overall, the results of the proposed methodologies show that they can bring substantial improvement to the task of automatic motion generation for visual quality inspection.
Ključne besede: robot learning, eobotic quality inspection, visual quality inspection
Objavljeno v DiRROS: 09.05.2024; Ogledov: 77; Prenosov: 368
.pdf Celotno besedilo (3,00 MB)
Gradivo ima več datotek! Več...

3.
4.
NutriGreen image dataset : a collection of annotated nutrition, organic, and vegan food products
Jan Drole, Igor Pravst, Tome Eftimov, Barbara Koroušić-Seljak, 2024, izvirni znanstveni članek

Povzetek: In this research, we introduce the NutriGreen dataset, which is a collection of images representing branded food products aimed for training segmentation models for detecting various labels on food packaging. Each image in the dataset comes with three distinct labels: one indicating its nutritional quality using the Nutri-Score, another denoting whether it is vegan or vegetarian origin with the V-label, and a third displaying the EU organic certification (BIO) logo.
Objavljeno v DiRROS: 23.04.2024; Ogledov: 125; Prenosov: 43
.pdf Celotno besedilo (2,84 MB)

5.
Towards deploying highly quantized neural networks on FPGA using chisel
Jure Vreča, Anton Biasizzo, 2023, objavljeni znanstveni prispevek na konferenci

Povzetek: We present chisel4ml, a Chisel-based tool that generates hardware for highly quantized neural networks described in QKeras. Such networks typically use parameters with bitwidths less than 8 bits and may have pruned connections. Chisel4ml can generate the highly quantized neural network as a single combinational circuit with pipeline registers in between the different layers. It supports heterogeneous quantization where each layer can have a different precision. The full parallelization enables very low-latency and high throughput inference, that are required for certain tasks. We illustrate this on the triggering system for the CERN Large Hadron Collider, which filters out events of interest and sends them on for further processing. We compare our tool against hls4ml, a high-level synthesis based approach for deploying similar neural networks. Chisel4ml is still under development. However, it already achieves comparable results to hls4ml for some neural network architectures. Chisel4ml is available on https://github.com/cs-jsi/chisel4ml.
Ključne besede: neural networks, QKeras, Chisel4ml
Objavljeno v DiRROS: 23.04.2024; Ogledov: 119; Prenosov: 64
.pdf Celotno besedilo (419,83 KB)
Gradivo ima več datotek! Več...

6.
How does day-to-day stress appraisal relate to coping among office workers in academia? : an ecological momentary assessment study
Stephanie Hulin, Larissa Bolliger, Junoš Lukan, Anneleen Caluwaerts, Rosalie De Neve, Mitja Luštrek, Dirk De Bacquer, Els Clays, 2023, izvirni znanstveni članek

Povzetek: Existing literature indicates that academic staff experience increasing levels of work stress. This study investigated associations between day-to-day threat and challenge appraisal and day-to-day problem-focused coping, emotion-focused coping, and seeking social support among academic office workers. This study is based on an Ecological Momentary Assessment (EMA) design with a 15-working day data collection period utilising our self-developed STRAW smartphone application. A total of 55 office workers from academic institutions in Belgium (n = 29) and Slovenia (n = 26) were included and 3665 item measurements were analysed. Participants were asked approximately every 90 min about their appraisal of stressful events (experienced during the working day) and their coping styles. For data analysis, we used an unstructured covariance matrix in our linear mixed models. Challenge appraisal predicted problem-focused coping and threat appraisal predicted emotion-focused coping. Our findings suggest an association between threat appraisal as well as challenge appraisal and seeking social support. Younger and female workers chose social support more often as a coping style. While working from home, participants were less likely to seek social support. The findings of our EMA study confirm previous research on the relationship between stress appraisal and coping with stress. Participants reported seeking social support less while working from home compared to working at the office, making the work location an aspect that deserves further research.
Ključne besede: academic setting, coping, work stress
Objavljeno v DiRROS: 25.03.2024; Ogledov: 142; Prenosov: 27
.pdf Celotno besedilo (366,59 KB)
Gradivo ima več datotek! Več...

7.
Hardware–software co-design of an audio feature extraction pipeline for machine learning applications
Jure Vreča, Ratko Pilipović, Anton Biasizzo, 2024, izvirni znanstveni članek

Povzetek: Keyword spotting is an important part of modern speech recognition pipelines. Typical contemporary keyword-spotting systems are based on Mel-Frequency Cepstral Coefficient (MFCC) audio features, which are relatively complex to compute. Considering the always-on nature of many keyword-spotting systems, it is prudent to optimize this part of the detection pipeline. We explore the simplifications of the MFCC audio features and derive a simplified version that can be more easily used in embedded applications. Additionally, we implement a hardware generator that generates an appropriate hardware pipeline for the simplified audio feature extraction. Using Chisel4ml framework, we integrate hardware generators into Python-based Keras framework, which facilitates the training process of the machine learning models using our simplified audio features.
Ključne besede: FPGA, MFCC, keyword spotting, chisel
Objavljeno v DiRROS: 25.03.2024; Ogledov: 321; Prenosov: 371
.pdf Celotno besedilo (1,05 MB)
Gradivo ima več datotek! Več...

8.
Comparing algorithm selection approaches on black-box optimization problems
Ana Kostovska, Anja Janković, Diederick Vermetten, Sašo Džeroski, Tome Eftimov, 2023, objavljeni znanstveni prispevek na konferenci

Objavljeno v DiRROS: 25.03.2024; Ogledov: 120; Prenosov: 42
.pdf Celotno besedilo (582,18 KB)

9.
10.
GPU adding-doubling algorithm for analysis of optical spectral images
Matija Milanič, Rok Hren, 2024, izvirni znanstveni članek

Ključne besede: medical imaging, medical optics, adding-doubling algorithm
Objavljeno v DiRROS: 12.03.2024; Ogledov: 209; Prenosov: 79
.pdf Celotno besedilo (1,05 MB)
Gradivo ima več datotek! Več...

Iskanje izvedeno v 0.64 sek.
Na vrh