GRUPPO DI INGEGNERIA INFORMATICA - ING-INF/05

The University of Padua, Department of Information Engineering, is seeking a postdoctoral researcher to contribute to a project funded by the CARIPARO foundation on turning Information Retrieval (IR) into a predictive science.

Predictability is a fundamental attribute of daily life: we expect familiar things to behave in familiar ways. In science, predictability has taken on more specific meanings; our understanding of a system, model, or method is validated by our ability to predict performance or outcomes, often in a quantified form. A particular challenge for the IR systems regarded here is that, ultimately, they create or affect a human experience.
The project will focus on 5 problem areas:

  1. Measures: We need a better understanding of the assumptions and user perceptions underlying different metrics, as a basis for judging about the differences between methods.
  2. Performance analysis: Instead of regarding only overall performance figures, we should develop rigorous and systematic evaluation protocols focused on explaining performance differences.
  3. Assumptions: The assumptions (and how much we depart from them) underlying our algorithms, evaluation methods, datasets, tasks, and measures should be identified and explicitly formulated.
  4. Application features: The gap between test collections and real-world applications should be reduced. Most importantly, we need to determine the features of datasets, systems, contexts, tasks that affect the performance of a system.
  5. Performance Models: We need to develop models of performance which describe how application features and assumptions affect the system performance in terms of the chosen measure, in order to leverage them for prediction of performance.

For more information on this topic, please visit: http://dx.doi.org/10.4230/DagMan.7.1.96

Specifically, the job requires:

  • Ph.D. in IR, Recommender Systems or Machine Learning (ML)
  • Knowledge of latest developments in IR and its experimental evaluation
  • Knowledge of latest developments in ML for IR
  • Good publication record in top-tier IR or ML conferences/journals
  • Strong development skills, preferably in Matlab, Python, Java

The successful candidate will join the Information Management Systems (IMS) group in Padua, working under the supervision of Professor Nicola Ferro.

The University of Padua, established in 1222, is one of the oldest universities in Europe, and one of the biggest in Italy. The University has 8 schools and 32 departments in almost all the disciplines and the research areas. There are more than: 2,100 research staff; 2,100 technical and administrative staff; 61,000 students; 11,700 graduate every year; and, 1,500 PhD students. The National Agency for the Evaluation of Universities and Research Institutes (ANVUR) ranked UNIPD as the first University in Italy for ICT in the Italian Research Assessment Exercise (VQR 2004-2010) and confirmed it in the subsequent one (VQR 2011-2014).

The Department of Information Engineering (DEI), established in 1987, is one of the biggest departments of the University of Padua, with about 100 research staff among full, associate, and assistant professors; 35 technical and administrative staff; about 130 PhD and Post-doc students; and, more than 2,000 graduate students. DEI has been the first department in Italy to be ISO 9001:2008 certified for quality. DEI has been judged in 2018 as one of the 100 Italian departments of excellence at a national level.

The Information Management Systems group (IMS), established in 1987, carries out research on many aspects of information processing, with a strong focus and a long history on information retrieval, experimental evaluation, multilingual information access, and digital libraries.

Salary will be 26,127 euros per annum. This post is full time with funding up to 22 months in the first instance.

Deadline for applications: 11 November 2019

For more information and application process, see:
https://www.dei.unipd.it/node/24657

Italian version:
https://www.dei.unipd.it/system/files/Bando%20n.%20109-2019%20assegno%20di%20ricerca%20Ferro.pdf

English version:
https://www.dei.unipd.it/system/files/Bando%20Assegno%20tipo%20A%29%20Ferro%20-%20inglese.pdf


Informal enquiries can be addressed to Professor Nicola Ferro (ferro@dei.unipd.it)