Nettsider med emneord «Machine Learning»

Christos Dimitrakakis talks about the challenge of identity protection.

João Gama, Associate Professor at the University of Porto, talks about the limitations of current machine learning and data mining algorithms in dealing with real-time data.

Water flow on, in and under glaciers still remains a poorly understood system. Water can flow over the surface of glaciers, as well as through channels inside and under the ice. The water thereby has strong influence both on ice melt and glacier dynamics, making glacial hydrology an important field of study.

Oppgaven utføres i samarbeid med Forsvarets forskningsinstitutt (FFI) på Kjeller. 

The use of lexical semantic information for the task of syntactic parsing has seen varied success. Recently, however, the use of lexical semantic clusters derived from large corpora has been shown to improve parsing performance. It is still unclear, however, how different properties of these clusters affect results. This project aims to investigate the use of different types of clusters during syntactic parsing. 

More precisely the idea is to use word clusters as a source for features in a statistical disambiguation model for a dependency parser. Generally, the clusters will group together words with similar distributional properties. The exact nature of these similarity relations, however, will vary depending on the types of context features that are used when performing the clustering. For this project we will basically be doing an extrinsic form of cluster evaluation then; investigating how different clustering parameters in turn affect the performance of a statistical parser. 

Classification models are often used to make decisions that affect humans: whether to approve a loan application, extend a job offer, or provide insurance. In such applications, it is desirable for the individuals to not only knowing the results but also have the ability to change the decisions of the model. For example, when a person is denied a loan by a credit scoring model, in addition to know why he/she can not received the loan, it is meaningful for the person to know what he/she can do to influence the decision, i.e. what are the input variables that, if values are changed, can alter the decision of the model. Otherwise, without this information, he/she will be denied the loan as long as the model is deployed, and – more importantly – will lack agency over a decision that affects their livelihood.

Ontologies are developed for representing requirements and specifications that can help domain experts and engineers do various tasks in the real world. Usually, as the ontologies are quite large, it takes a long time for human users to understand the ontologies and it is also quite difficult for machines to do reasoning on such large ontologies. The goal is that we can explore different machine-learning techniques to extract relevant parts of the ontology and reduce reasoning time. 

Two new papers from SANT were presented at the NoDaLiDa conference.

In the field of robotics, creating grippers that can perform tasks similar to the capabilities of a human hand, is still a challenge yet to be solved. 

The Tactile Internet –a communication network that is capable of delivering real-time control, touch, and sensing/actuation information through sufficiently reliable, responsive, and intelligent connectivity – is revolutionizing the understanding of what is possible through wireless communication systems, pushing boundaries of Internet-based applications to remote physical interaction. Such remote interaction capability can be used in surgery, driving, drone-based transportation, immersive education, and adventure, etc. Although the community envisions the bright future of the tactile internet, few works implement the physical platform and specify the blockages in enabling a smooth quality of experience (QoE) during the interaction. In this project, the candidate(s) will work in a team to identify those blockages and propose the novel algorithm in network stack to enable the avatar in the real world.

Obstructive sleep apnea (OSA) is a common but severely under-diagnosed sleep disorder that affects the natural breathing cycle during sleep with the periods of reduced respiration or no airflow at all. It is our long-term goal to increase the percentage of diagnosed OSA cases, reduce the time to diagnosis, and support long term monitoring of patients with user friendly and cost-efficient tools for sleep analysis at home. Core elements are mobile computing platforms (e.g., smartphones), consumer electronics sensors, and machine learning for OSA detection.

This year's Hassel lecture is headed by Professor Veronique Van Speybroeck. The first day Professor Speybroeck will present the lecture "Operando modeling of functional nanostructured materials for sustainable chemistry, nanosensing and clean energy."

This year's Hassel lecture is headed by Professor Veronique Van Speybroeck. The second day Professor Speybroeck will present the lecture "From quantum mechanics to machine learning: Bridging length and time scales in modeling nanoporous materials at operating conditions."

This year's Hassel lecture is headed by Professor Veronique Van Speybroeck. The second day Professor Speybroeck will present the lecture "From quantum mechanics to machine learning: Bridging length and time scales in modeling nanoporous materials at operating conditions."

Neural network architectures currently used for histopathology were developed for natural images (not medical ones). Some of these architectures have been shown to have inductive biases for natural images, i.e., the neural network's architecture has some information about the natural images. These biases can be very helpful for processing natural images. This project aims to find architectures with an inductive bias for histopathological images. We will do it in an automated way (using neural architecture search) instead of trying to build an architecture manually.