Nettsider med emneord «Deep learning/neural networks»

The goal of the project is to co-develop technology and proposals for regulatory measures to reduce vulnerabilities regarding robotics.

PIRC targets a psychology-inspired computing breakthrough through research combining insight from cognitive psychology with computational intelligence to build models that forecast future events and respond dynamically.

The main objective of this work is to improve the utility of new small satellites for Earth Observation (EO), by researching machine learning techniques to obtain improved and useful detection, classification, and identification capabilities from space.

Devices, like smart-watches, that can collect health data from "everybody" all the time, and machine learning (ML) to analyze this data will strongly impact future health solutions.

The adaptive immune system records all past and ongoing battles with disease and infection in the form of immune memory, stored in the form of DNA of immune receptors of adaptive immune cells. However, deciphering these signals is a grand challenge of immunology, requiring sophisticated machine learning.

Accurate mapping of surface greenhouse gas fluxes is necessary for the validation and calibration of climate models. In this project, we develop a novel framework using drone observations and machine learning to estimate greenhouse gas fluxes at a regional scale.

In MASSIVE, the project team aims at improving glacier mapping and surface glacier mass balance estimation techniques with the help of machine learning, especially deep learning. We will develop the methodology for glaciers in Norway, Svalbard, the European Alps and the Himalayas and then expand it to regions with different glacier characteristics.

LATICE aims to advance the knowledge base concerning land atmosphere interactions through improved model representation of snow, permafrost, hydrology and large-scale vegetation processes representative of high latitudes, including; 1) New ground observations (gap filling using ML); 2) Land surface model parametrization using data science methods; 3) Seasonal snow cover dynamics using data assimilation and Earth observations (e.g. satellite data, drones).

The CryoGrid community model is a flexible toolbox for simulating the ground thermal regime and the ice/water balance for permafrost and glaciers. The CryoGrid community model can accommodate a wide variety of application scenarios, which is achieved by fully modular structures through object-oriented programming.

Exploring the fundamental constituents of the Universe physicists are faced with very serious calculational bottlenecks. To compare new physics models to data we need to perform very computationally expensive calculations in quantum field theory (QFT).

G-protein coupled receptor (GPCRs) form the largest superfamily of membrane proteins in human. 34% of the marketed small molecule drugs bind to GPCRs. Tens of millions of compounds are commercially available for screening against GPCRs in experimental setting, which is impractical for academia and industry.

For maritime safety surveillance we develop new approaches
based on the availability of large arrays of sensors, which
monitor condition and performance of vessels, machinery, or
power systems.

AI, statistical models and machine learning methods can often be seen as black boxes to those who construct the model and/or to those who use or are exposed to the methods.

BigInsight produces innovative solutions for key data-driven challenges facing our consortium of private, public and research partners, by developing original statistical and machine learning methodologies.

We develop and apply methods based on machine learning for chemistry and materials science. At the method level, our focus is on data (datasets computed with quantum mechanics methods), representations (graphs based on electronic structure theory), and models (graph neural networks and boosted trees).

SPATUS is a Thematic Research Group funded by UiO:Energy.

Bayesian methods have recently regained a significant amount of attention in the machine community due to the development of scalable approximate Bayesian inference techniques. There are several advantages of using Bayesian approach: Parameter and prediction uncertainty become easily available, facilitating rigid statistical analysis. Furthermore, prior knowledge can be incorporated.