Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 4

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  skupienie
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
1
Content available remote Measuring Trustworthiness in Neuro-Symbolic Integration
EN
Neuro-symbolic integration of symbolic and subsymbolic techniques represents a fast-growing AI trend aimed at mitigating the issues of neural networks in terms of decision processes, reasoning, and interpretability. Several state-of-the-art neuro-symbolic approaches aim at improving performance, most of them focusing on proving their effectiveness in terms of raw predictive performance and/or reasoning capabilities. Meanwhile, few efforts have been devoted to increasing model trustworthiness, interpretability, and efficiency - mostly due to the complexity of measuring effectively improvements in terms of trustworthiness and interpretability. This is why here we analyse and discuss the need for ad-hoc trustworthiness metrics for neuro-symbolic techniques. We focus on two popular paradigms mixing subsymbolic computation and symbolic knowledge, namely: (i) symbolic knowledge extraction (SKE), aimed at mapping subsymbolic models into human-interpretable knowledge bases; and (ii) symbolic knowledge injection (SKI), aimed at forcing subsymbolic models to adhere to a given symbolic knowledge. We first emphasise the need for assessing neuro-symbolic approaches from a trustworthiness perspective, highlighting the research challenges linked with this evaluation and the need for ad-hoc trust definitions. Then we summarise recent developments in SKE and SKI metrics focusing specifically on several trustworthiness pillars such as interpretability, efficiency, and robustness of neuro-symbolic methods. Finally, we highlight open research opportunities towards reliable and flexible trustworthiness metrics for neuro-symbolic integration.
2
Content available remote Fast focusing iterative migration of magnetic anomalies
EN
Magnetic anomalies are interpreted for subsurface geological information. Three-dimensional inversion of magnetic data is a challenging quantitative approach for interpreting the data. The rapid iterative migration technique could be a good and fast alternative for the inversion method. The focusing iterative migration that employs a focusing stabilizer can generate focused migration models which justify geological interfaces, adequately. This paper introduces a new algorithm using a relaxed steepest descent method and a sigmoid stabilizer for fast focusing migration of magnetic felds. The developed method can improve the computational efciency of focusing iterative migration by reducing the required iterations. The better performance of the new method is demonstrated by two numerical models and a real case study. The magnetic anomaly over San Nicolas massive sulfde deposit in Mexico is used for the case study. Compared with the drilling information, the iterative migration methods produce robust migration models for the San Nicolas deposit.
EN
Heat waves were analysed on the basis of maximum daily temperature from 24 meteorological stations in Poland. Heat waves are defined as the longest continuous period during which Tmax (daily maximum air temperature) is equal to or higher than 30°C in at least three days, the mean Tmax during the whole heat wave is equal or higher than 30°C and Tmax does not drop below 25°C during the whole period of heat wave duration. Heat waves occur in Poland from April to September with their maximums in July and August. Four-day-long heat waves are most frequent but the longest one lasted 31 days. The most persistent heat waves were in 1994 and 2015. An increasing trend in heat wave frequency and intensity is observed in Poland, however the increase is statistically significant at only about 60% of analysed stations. Four synoptic patterns favouring heat waves have been distinguished. A strong high over the Azores accompanies all of them, as well as slightly higher than normal pressure over Central Europe – this causes calm and sunny weather over Poland. Strong blocking appears over the North Atlantic during heat wave events, proofing that the development of strong heat waves in Poland is related to large scale circulation and that they are not of local origin. The analysis of the impact of soil moisture in months leading up to the development of heat waves should be the next step in analysis.
4
Content available remote Identifying the role of initial wave parameters on tsunami focusing
EN
Unexpected local tsunami amplification, which is referred to as tsunami focusing, is attributed to two different mechanisms: bathymetric features of the ocean bottom such as underwater ridges and dipolar shape of the initial wave itself. In this study, we characterize the latter; that is, we explore how amplitude and location of the focusing point vary with certain geometric parameters of the initial wave such as its steepness and crest length. Our results reveal two important features of tsunami focusing: for mild waves maximum wave amplitude increases significantly with transverse length of wave crest, while location of the focusing point is almost invariant. For steep waves, on the other hand, increasing crest length dislocates focusing point significantly, while it causes a rather small increase in wave maximum.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.