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: "ključne besede" (regression) .

1 - 3 / 3
Na začetekNa prejšnjo stran1Na naslednjo stranNa konec
1.
2.
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: 362; Prenosov: 438
.pdf Celotno besedilo (2,49 MB)
Gradivo ima več datotek! Več...

3.
Modelling dominant tree heights of Fagus sylvatica L. using function-on-scalar regression based on forest inventory data
Markus Engel, Tobias Mette, Wolfgang Falk, Werner Poschenrieder, Jonas Fridman, Mitja Skudnik, 2023, izvirni znanstveni članek

Povzetek: European beech (Fagus sylvatica L.) is an important tree species throughout Europe but shifts in its suitable habitats are expected in the future due to climate change. Finding provenances that are still economically viable and ecologically resilient is an ongoing field of research. We modelled the dominant tree heights of European beech as a trait reflecting growth performance dependent on provenance, climate and soil conditions. We derived dominant tree heights from national forest inventory (NFI) data from six European countries spanning over large ecological gradients. We performed function-on-scalar regression using hierarchical generalized additive models (HGAM) to model both the global effects shared among all provenances and the effects specific to a particular provenance. By comparing predictions for a reference period of 1981–2010 and 2071–2100 in a RCP 8.5 scenario, we showed that changes in growth performance can be expected in the future. Dominant tree heights decreased in Southern and Central Europe but increased in Northern Europe by more than 10 m. Changes in growth performance were always accompanied by a change in beech provenances, assuming assisted migration without dispersal limitations. Our results support the concept of assisted migration for the building of resilient future forests and emphasize the use of genetic data for future growth predictions.
Ključne besede: hierarchical GAMs, functional regression, Fagus sylvatica, provenance, assisted migration
Objavljeno v DiRROS: 21.03.2023; Ogledov: 777; Prenosov: 356
.pdf Celotno besedilo (3,65 MB)
Gradivo ima več datotek! Več...

Iskanje izvedeno v 0.1 sek.
Na vrh