Research Position in Doctoral Network "LEMUR"
The newly established EU Doctoral Network (DN) 'Learning with Multiple Representations (LEMUR)' is concerned with machine learning methods that specifically use different representations (such as symbolic and sub-symbolic representations) in a model to make concrete machine learning tasks more efficient, robust and secure. The position holder (so-called Early Stage Researchers (ESRs)) in the LEMUR project will work on the task of how such methods can be developed or improved, what mathematical properties they possess and how they can be profitably used in practice. In the ESR4 project, which is part of the machine learning group at Bielefeld University, the focus is on discriminative dimensionality reduction methods which are capable of dealing with auxiliary information. These can be used to visualize and interactively explore machine learning models including deep networks.
The tasks essentially comprise:
- development and implementation of new dimensionality reduction methods to represent the function of complex learning architectures (35 %)
- implementation for interactive scenarios, e.g. for data analysis or observation of training progress (30 %)
- application in cooperation with secondment partners of the DN (15 %)
- participation in DN events, such as summer schools (10 %)
- cooperation with international project partners (10 %)
An important component of the DN are secondments to project partners in industry and research. This offers the opportunity to gain a deeper insight into methods used and a broader view of applications.
Since the position is financed by third-party funds, the following must be observed according to the requirements of the third-party funder: Due to the objective of DNs to strengthen international cooperation, applicants may not have been resident and/or professionally active in Germany for more than twelve months during the last three years at the time of recruitment. Furthermore, applicants should not have completed their university degree more than four years ago at the time of recruitment. It is a full-time position; a reduction in working-hours is only possible in compliance with the regulations of the third-party funder, for example for parental leave.
Employment is conducive to scientific qualification.
- relevant university degree in computer sciences, mathematics or technically equivalent fields of research
- very good programming skills, especially Python
- knowledge of machine learning
- independent, conscientious and diligent working style
- cooperative and team-oriented working style
- openness to application-oriented questions in an industrial context
- fluent written and oral English language skills
- knowledge of dimensionality reduction methods
- experience in deep learning
- advanced skills in mathematical modeling
- salary according to Remuneration level 13 TV-L (min. 4.188,38 EUR)
- befristet (3 years) (§ 2 (1) sentence 1 of the WissZeitVG; in accordance with the provisions of the WissZeitVG and the Agreement on Satisfactory Conditions of Employment, the length of contract may differ in individual cases)
- Vollzeit
- internal and external training opportunities
- variety of health, consulting and prevention services
- reconcilability of family and working life
- flexible working hours
- job ticket for regional public transport network, good transport connection
- supplementary company pension
- collegial working environment
- open and pleasant working atmosphere
- exciting, varied tasks
- modern work environment with digital processes
We are looking forward to receiving your application. For full consideration, your application should be received via either
email (a single PDF document is required) sent to bhammer@techfak.uni-bielefeld.de or post (see postal address). Please mark your application with the identification code: Wiss23103. Please note that the possibility of privacy breaches and unauthorized access by third parties cannot be excluded when communicating via unencrypted e-mail. For Information on the processing of personal data
click here.
application deadline: 12.04.2023