Predicting Requests in Large-Scale P2P Ridesharing

Peer-to-peer ridesharing (P2P-RS) enables people to arrange one-time rides with their own private cars, without the involvement of professional drivers. It is a prominent collective intelligence application producing significant benefits both for individuals (reduced costs) and for the entire community (reduced pollution and traffic), as we showed in a recent publication where we proposed an online approximate solution algorithm for large-scale P2P-RS [Bistaffa et al., “A Computational Approach to Quantify the Benefits of Ridesharing for Policy Makers and Travellers”, IEEE T-ITS].

Intelligent SpatiaL Audio (ISLA)

The ISLA project is aimed at producing a step-change in the way that audio-related applications are conceived, adding ambient intelligence to spatial audio synthesis and monitoring tasks. This may be possible in the new Internet of Things (IoT) era, where multiple sensoring devices with high-performance computing capabilities are interconnected.

NeuroCONVO: Brain Waves with Neural Networks

The brain generates activity in the form of oscillations. Brain waves span from very slow rhythms, typical of sleep, to faster oscillations during attention and cognitive processing. Moreover, changes of brain oscillations are markers of some neurological diseases. Given the dynamism of brain activity, these events are far from stationary and thus their identification in real time is a daunting task.

Anomaly Awareness

Algorithms that detect anomalies have to learn normal behaviour to be able to identify anomalous behaviour. If broad features of the expected anomalies are known the use of supervised Machine Learning (ML) is in order. But by definition, the most interesting anomalies are those unexpected, and in that case, unsupervised ML should be used. However, unsupervised strategies are substantially less powerful than possible supervised methods –a catch 22 situation.


Machine learning is one of the keys in the development of modern Earth Observation satellite missions. Model training requires precise Earth simulation as basis of applications such as vegetation monitoring, prospecting for minerals, soil use and climate change studies, among others.

Dark Machines

Dark Machines ( is an initiative to develop and apply machine learning methods to accelerate dark matter searches. It is composed of more than 300 high-energy physicists, astroparticle physicists and astrophysicists, from theory and experiment, as well as computer scientists.

Machine Learning in Magnetic Resonance

Low back pain (LBP) is a very prevalent pathology and a frequent cause of disability. It is associated with rising costs for the health system and for society in developed countries, affecting 70% of the general population at some time in their lives, with an annual incidence of 40%

The multidisciplinary group lead by María de la Iglesia-Vayá from the Prince Felipe Research Center (CIPF) uses Artemisa to develop the first massive and open-access data repository of lumbar MRI for International collaborative research.




Artemisa is a High Performance Computing Facility oriented to Machine
Learning and Artificial Intelligence research assisted by GPUs coprocessors.

It uses state-of-the-art CPUs and GPUs to allow computing
projects to develop and run their most advanced algorithms.

The facility is composed by two user interface machines in which they
can develop and tests their workflow and 23 machines to send batch
jobs. All batch machines contain an NVIDIA GPU Volta V100 to
assist with their AI algorithms.

NVIDIA Tesla Volta V100 SXM2

The Artemisa facility provides one server with 4 GPUs NVIDIA Tesla Volta V100
SMX2 with 32 GB memory each. This GPU model communicates
using the NVLink, which provides bi-sectional Bandwidth up to 300 GB/s.
Each GPU provides 15 TFlops for single precision (32 bits)
and 7 TFlops for double precision. Furthermore, those GPU provides
specific tensor computing capacity with a rate of 128 TFlops on
single precision operations.