CRYSTALLINE PhD position - Efficient Algorithms and Accelerator Architectures for Distributed Edge-AI Systems (f/m/d)

19 Jan 2024
Job Information

Organisation/Company

Silicon Austria Labs

Department

Human Resources

Research Field

Engineering » Computer engineering
Engineering » Electronic engineering
Technology » Computer technology
Technology » Internet technology
Computer science » Computer architecture
Computer science » Computer hardware
Computer science » Computer systems
Computer science » Informatics
Computer science » Programming
Computer science » Systems design

Researcher Profile

First Stage Researcher (R1)

Country

Austria

Application Deadline

29 Mar 2024 - 12:22 (Europe/Vienna)

Type of Contract

Temporary

Job Status

Full-time

Hours Per Week

38.5

Offer Starting Date

1 Jul 2024

Is the job funded through the EU Research Framework Programme?

HE / MSCA COFUND

Is the Job related to staff position within a Research Infrastructure?

No

Offer Description

In the "Efficient Algorithms and Accelerator Architectures for Distributed Edge AI Systems", you will explore the world of distributed, decentralized and federated AI systems. These cutting-edge systems enable efficient knowledge sharing between autonomous edge devices, revolutionizing their operational capabilities. This PhD position focuses on exploring space- and energy-efficient edge AI tailored for distributed systems. In contrast to prevailing approaches that focus primarily on inference, this work aims to comprehensively cover the training and fine-tuning aspects of AI.
 

• Design and implementation of innovative distributed AI methods and algorithms. 
• Customizing these methods to unique constraints of power- and resource-limited environments of edge devices and networks.  
• Investigate novel accelerator architectures for embedded AI applications, tailoring designs to maximize both performance and energy efficiency. 
• Explore the potential for using quantization methods particular attention to the implications for training and fine-tuning neural networks on edge devices. 
• Investigate the reliability and resilience of such systems with a particular focus on the influence of the chosen acceleration and quantization scheme. 
• Systematically benchmark the designs against existing solutions and evaluate their performance in a variety of use cases. Provide a comparative analysis of energy efficiency, speed and accuracy and demonstrate the competitiveness of the proposed solutions.  
• Mentor and guide Masters students, supporting their academic and professional development.  
• Contribute to the scientific community by publishing of your research in high-impact journals and presenting your findings at international conferences.

Requirements

Research Field

All

Education Level

Master Degree or equivalent

Skills/Qualifications

• Master's degree in computer science, applied mathematics, robotics, cyber-physical systems, data science, electrical engineering or a related field.  
• Strong programming and algorithmic skills. 
• Knowledge of hardware description languages is an advantage. 
• Experience in machine learning, including deep learning (LSTMs, CNNs, transformers, ...); specific experience in distributed computing systems (fog/cloud computing) and edge systems is beneficial. 
• Good communication skills, including basic presentation and scientific writing skills. 
• Excellent written and oral communication skills in English.
• Enthusiasm for developing new ideas and a positive attitude towards new challenges. 
• Ability to work independently, be well organised, produce high quality documents and meet deadlines. 
• Project experience and/or publications in related fields are beneficial.

Languages

ENGLISH

Level

Excellent

Additional Information
Selection process

In order for a JS candidate to be eligible for CRYSTALLINE, she or he has to comply with the MSCA-COFUND Doctoral Program (DP) eligibility rules, specific CRYSTALLINE application requirements, and with fundamental ethic principles.

The MSCA-COFUND DP rules require
researchers to (i) be doctoral candidates, i.e., not already in possession of a doctoral degree at the call deadline (whereas researchers who have successfully defended their doctoral thesis but who have not yet formally been awarded the doctoral degree will not be considered eligible), and to (ii) meet the MSCA mobility rule, i.e., have not resided and/or carried out their main activity (work, studies, etc.) in Austria for more than 12 months in the 3 years immediately before the call deadline (time spent as part of a procedure for obtaining refugee status under the Geneva Convention is not taken into account).

The CRYSTALLINE application requirements are met if an applicant (i) holds a master’s degree in a technical discipline, (ii) provides a complete set of application documents, and (iii) complies with the submission rules for the application documents as laid out at the CRYSTALLINE website.

Work Location(s)

Number of offers available

9

Company/Institute

Silicon Austria Labs GmbH

Country

Austria

Geofield

Where to apply

Website

https://research-network.silicon-austria.com/about-crystalline/

Contact

State/Province

Steiermark

City

Graz

Website

https://silicon-austria-labs.com

Street

Sandgasse 34

Postal Code

8010

E-Mail

[email protected]

STATUS: EXPIRED