JOB DESCRIPTION

Closed for Applications

Title

Early Stage Researcher (ESR9)  

PhD fellowship in deep learning model development to predict cheese properties.

Project Title

“E-MUSE Complex microbial ecosystems multiscale modelling: mechanistic and data driven approaches integration” MSCA-ITN-2020 European Training Network

Hosting Organization

University of Szeged (USZ), established in Dugonics Tér 13, Szeged 6720, Hungary

Researcher Profiles

ESRs

Research Field

Machine learning on multi-omics data, feature selection, support mechanistic models with machine learning

Application Deadline

31.05.2021, 23:59;  prolonged till 18.07. 2021 23:59 - Europe/Brussels

Envisaged Job Starting Date

01.09.2021 

Duration

The appointment will initially be for 1 year. After a satisfactory evaluation of the initial appointment, the contract will be extended for a duration of 2 years.

Contract Type

full-time employment (based on COVID-19 evolution and restrictions, possibility to start remotely, once situation allows the presence is required)

Remuneration

€2 530.98 Eur gross salary/month

+ €600 gross mobility allowance/month and €500 gross family allowance/month (if applicable)

Taxation and Social (including Pension) Contribution deductions based on National and company regulations will apply.

Objectives

The objective of this ESR project is to train young researchers with strong programming and/or data scientist background to tackle machine and deep learning problems in the field of predictive modelling and multi-omics. As an example, the ESR will develop deep learning models using multi-omics data to estimate product properties at higher levels (population and/or macroscopic  level). At the same time models will be trained to provide a priori knowledge for network-based approaches. For this purpose, ESR will explore novel neural network architectures and learning approaches. This includes experiments with different model architectures such as recurrent, convolutional (e.g. for the determination of protein-protein interactions), attention-based networks. Graph neural networks are recently proposed in systems biology, where a weighted graph is a commonly used representation. Ensemble learning embraces the decision of several machine learning models. We plan to experiment with the recent idea of ensemble of ensembles combined with unsupervised methods. Multi-task learning is a promising way to improve model performance where different learning tasks need similar input data, since this improves the generalization capability of models. This fits to the case of different biomarkers identification based on multi-omics data.

Expected Results

We expect to deliver tools (open source program) and methods (deep learning models, data manipulation and learning techniques) by which we can estimate the high level properties (e.g. taste, texture) of cheese product, based on lower level predictors.

Planned Secondments

In total, 2 months will be spent at Alma Mater Studorium – Universita di Bologna (UNIBO) in Italy (network theory of multi-omics data).

Enrolment in Doctoral degree

University of Szeged (USZ) (https://u-szeged.hu/english)

Requirements

Required Education Level

Master’s degree in computer science or equivalent degrees ideally with a strong background in data science, machine learning, deep learning no later than September 2021. You should NOT have any kind of PhD degree. Previous research experience (which must be no longer than 4 years) although appreciated, is not mandatory. 

Skills / Qualifications

  • Computer science, data science, machine learning, deep learning background

  • Provable software development skills in one of common programming languages (e.g. Java, C++)

  • Python, Tensorflow, PyTorch experience is favoured

  • Willingness to learn the necessary biology background in order to work with biological data properly

  • Educational background and previous research experience relevant for the chosen position

  • Networking and good communications skills (writing and presentation skills)

  • Willingness to travel abroad for the purpose of research, training and dissemination

Specific Requirements

For the eligibility please check: Eligibility Criteria

Required Languages

English: B2, good oral and written communication skills in English are compulsory

Supervisors Team

The Univeristy of Szeged has vast knowledge and experience in various fields of machine learning and deep learning. The  Research Group on Artificial Intelligence, at which the proposed work will be carried out mainly focuses on image, voice and text based AI research, including medical AI applications and cutting-edge AI solutions for international industry partners.


The lead supervisor is Dr. László Vidács, senior researcher and expert in Machine Learning, Artificial Intelligence and Software at the Department of Software Engineering, University of Szeged in Hungary. He is also the deputy head of  MTA-SZTE Research Group on Artificial Intelligence with 6 researchers with PhD, 4 PhD students and 8 young researchers and has a leading role in 11 EU funded and industrial R&D projects. He is a supervisor of 4 ongoing PhD, 6 Masters students.


The co-supervisor is Prof. Daniel Remondini - the director of the Applied Physics and Systems Biophysics Laboratory, Department of Physics and Astronomy (DIFA) of the Alma Mater Studiorum - University of Bologna, Italy. He has also specific expertises on biomedical data analysis (Machine Learning, Deep Learning), complex network theory and its applications to BioMedicine. He is supervisor of 8 PhD students in Physics (5 ongoing, 2 with a PostDoc position abroad and 1 with a Research Assistant position in Bologna), 1 ITN PhD student, 3 PostDoc students, and of >50 Undergraduate and Master Thesis students in Physics.

Host Institution Description 

The University of Szeged has been one of the best universities in Hungary for years, according to the international QS World University Rankings. Since Hungary is a member of the European Higher Education Area (EHEA), it also has EU accreditation. It is a research university with 12 Faculties, covering 700 research areas at 19 Doctoral Schools. Outstanding professors have worked at the university, including the Nobel Laureate Albert Szent-Györgyi (1937), who was the first to isolate Vitamin C, extracting it from Szeged paprika. Our student body grew to 21000, in which the number of international students exceeds 4000 coming from 115 countries. Artificial Intelligence research is mainly concentrated in three research units: the Department of Software Engineering, the Department of Algorithms and Artificial Intelligence and the Research Group on Artificial Intelligence.


Apart from quality education these groups have also been doing high impact academic research on various fields of AI including speech technology, natural language processing, software engineering, security, deep learning, self-organizing systems and theory and methodology of machine learning. Applications include several healthcare domains where state-of-the-art NLP and deep learning based image processing technologies are applied. USZ has active research in analyzing deep learning algorithms and examining the adversarial robustness of machine learning algorithms.
The University is located in the sunniest city of Hungary attracting thousands of young people due to its lively, urban lifestyle and colourful festivals. More information on student’s life in Szeged:


•    Welcome to the University of Szeged – The First Impressions
(https://www.youtube.com/watch?v=gsnDg8rWe6w)
•    Education at the University of Szeged (https://www.youtube.com/watch?v=ZCUXu4qEEfk)
•    Information for Prospective and Newly Admitted Students
(https://u-szeged.hu/english/currentstudents)
•    University of Szeged
(https://apply.u-szeged.hu/en_GB/institutions/institution/1-university-szeged)