EU funding for SNE Lab project Deep Learning
Dr. Andy Pimentel, scientist at the System and Network Engineering Lab (SNEL) at the Informatics Institute of the University of Amsterdam has received a €350.000 grant as partner of the ALOHA project. The project develops a software framework for runtime-Adaptive and secure Deep Learning On Heterogeneous Architectures.
Deep Learning algorithms a promising instrument
Deep Learning (DL) algorithms are a promising instrument in artificial intelligence, achieving very high performance in numerous recognition, identification, and classification tasks. To foster the pervasive adoption of DL in a vast scope of new applications and markets, a step forward is needed towards the implementation of the on-line classification task (called inference) on low-power embedded systems, enabling a shift to the edge computing paradigm.
Nevertheless, when DL is moved at the edge, severe performance requirements must coexist with tight constraints in terms of power/energy consumption, posing the need for parallel and energy-efficient heterogeneous computing platforms.
Deployment of Deep Learning unaffordable?
Unfortunately, programming for this kind of architectures requires advanced skills and significant effort, also considering that Deep Learning algorithms are designed to improve precision, without considering the limitations of the device that will execute the inference. Thus, the deployment of Deep Learning algorithms on heterogeneous architectures is often unaffordable for SMEs and midcaps without adequate support from software development tools.
ALOHA: the solution
Project ALOHA will facilitate the implementation of Deep Learning on heterogeneous low-energy computing platforms. The project will develop a software development tool flow, automating: algorithm design and analysis; porting of the inference tasks to heterogeneous embedded architectures, with optimized mapping and scheduling; implementation of middleware and primitives controlling the target platform, to optimize power and energy savings.
During the development of the ALOHA tool flow, several main features will be addressed, such as architecture-awareness (the features of the embedded architecture will be considered starting from the algorithm design), adaptivity, security, productivity, and extensibility. ALOHA will be assessed over three different use-cases, involving surveillance, smart industry automation, and medical application domains.
The ALOHA consortium exits of fourteen partners, has a budget of six million euro’s and will run from 2017 until 2021. Partners are: STMicroelectronics srl (IT), Università di Cagliari (IT), University of Amsterdam (NL), Leiden University (NL), ETH Zürich (CH), Università degli Studi di Sassari (IT), PKE Electronics AG (AT), CA Technologies Development Spain SA (ES), Software Competence Center Hagenberg GmbH (AT), Santer Reply Spa (IT), IBM israel – science and technology ltd (IL), Systhmata Ypologistikis Orashs Irida Labs ae (IT), Pluribus One srl (IT), MedyMatch Technology, ltd.(IL).
Associate Professor Andy Pimentel leads the Parallel Computing Systems (PCS) group within the System and Network Engineering Lab. The PCS group performs research on the design, programming and run-time management of multi-core and multi-processor computer systems. Andy has a master’s and PhD’s degree in Computer Science at the University of Amsterdam.