EU Projects Office

The focus of SMARTHEP is a central question in a data-rich environment: how to make the most of the available data to take decisions fast and efficiently, making the most of the available data. The main purpose of SMARTHEP is to train a new generation of inter-sector researchers and give them the tools to tackle this challenge, by processing large datasets in real- time, aided by Machine Learning and hybrid computing architectures. The results of SMARTHEP will benefit the HEP community in providing cutting edge technology and algorithms for the area of data selection (triggering) and particle detection, leading to precise measurement of the fundamental constituents of matter and enabling the discovery of new physics processes.

Coordinator: ULUND, Sweden

Scientist in Charge from CERN:
Monica Pepe-Altarelli

Full costs of the project: 3.2 M€

EU funding: 3.2 M€

EU funding for CERN: 281 k€

RAISE

RAISE: AI- and Simulation-Based Engineering at Exascale

Compute- and data-driven research encompasses a broad spectrum of disciplines and is the key to Europe’s global success in various scientific and economic fields. The massive amount of data produced by such technologies demands novel methods to post-process, analyze, and to reveal valuable mechanisms. The development of artificial intelligence (AI) methods is rapidly proceeding and they are progressively applied to many stages of workflows to solve complex problems. Analyzing and processing big data require high computational power and scalable AI solutions. Therefore, it becomes mandatory to develop entirely new workflows from current applications that efficiently run on future high-performance computing architectures at Exascale. The Center of Excellence for AI- and Simulation-based Engineering at Exascale (AISee) will be the excellent enabler for the advancement of such technologies in Europe on industrial and academic levels, and a driver for novel intertwined AI and HPC methods. These technologies will be advanced along representative use-cases, covering a wide spectrum of academic and industrial applications, e.g., coming from wind energy harvesting, wetting hydrodynamics, manufacturing, physics, turbomachinery, and aerospace.

Coordinator: FZJ, Germany

Scientist in Charge from CERN: Maria Girone

Full costs of the project: 4.9 M€

EU funding: 4.9 M€

EU funding for CERN: 517 K€

1 January 2021 - 31 December 2023

hls4ml

hls4ml: High Level Synthesis for Machine Learning

With Deep Learning becoming ubiquitous in our life, running Deep Learning algorithms in real time on an heterogeneous set of hardware platforms is a pressing need in many aspects of our society. While traditional workflows based on standard CPUs and GPUs are established, Deep Learning inference on low-power devices (e.g., cars, smart phones, watches, etc) is gaining more attention. Typically, this would require strong background in electronic engineering to convert a neural network into a Digital Signal Processor. hls4ml proposes to develop a complete open-software library to automatically convert Deep Neural Networks to electronic circuits, using High Level Synthesis tools. With a large basis of potential applications (e.g., autonomous cars, medical devices, portable monitoring devices, custom electronics as in the real-time data processing system of large-scale scientific experiments, etc.), the hls4ml library would assists users by automatising the logic circuit design as well as by reducing resource utilisation while preserving accuracy.

Coordinator: CERN, Switzerland

Scientist in Charge from CERN:
Maurizio Pierini

Full costs of the project: 150 k€

EU funding: 150 k€

EU funding for CERN: 150 k€

1 April 2021 - 30 September 2022

Gamma MRI

Gamma MRI: the future of molecular imaging

Gamma-MRI will develop a clinical molecular imaging device based on the physical principle of anisotropic gamma emission from hyperpolarised metastable xenon. Gamma-MRI is a game-changer imaging technology, combining the high sensitivity of gamma ray detection and the high resolution and flexibility of MRI, bringing down by multiple fold the cost of molecular imaging. Six closely interlinked work packages will cover: production of hyperpolarised gamma-emitting xenon isomers; preserving hyperpolarisation until delivery to targeted organ; developing advanced image acquisition and reconstruction using physics- and artificial intelligence- based approaches; designing and assembling the prototype upon a low field versatile magnet; and implementing the first preclinical Gamma-MRI brain imaging experiment.

Coordinator: HES-SO, Switzerland

Scientist in Charge from CERN:
Magdalena Kowalska

Full costs of the project: 3.3 M€

EU funding: 3.3 M€

EU funding for CERN: 243 k€

1 April 2021 - 31 March 2024

Subscribe to