Attila Kertesz-Farkas

Question Answering (QA) and Machine Reasoning (MR) have become a crucial application problem in evaluating the progress of AI systems in the realm of natural language processing and understanding, and to measure the progress of machine intelligence in general. However, most of the advances have focused on “shallow” QA tasks that can be tackled very effectively by existing retrieval-based techniques. Deep learning-based methods achieve human-like performance on benchmark datasets; however, it is suspected that these methods merely learn to match answers to questions or focus attention on specific words and pieces of text and they do not perform real reasoning or cognition.

There are several research projects related to this topic on BSc, MSc, PhD and postdoc levels ranging from testing a scrutinizing recent methods to developing your own method for machine reasoning using augmented machines (RNNs with external memory which is learned to be used by data, also known as neural Turing machines).

All projects are conducted in English, so it’s a good opportunity to improve your communication & presentation skills in English. All projects involve programming.

Mass spectrometry has become the de facto method to identify molecules in complex mixtures (e.g. blood, cell, food) in many areas. A mass spectrometer analyses a complex mixture of biological or chemical samples and produces tens of thousands of spectra from the input sample. These spectrum data can be considered as fingerprints of the samples and the main computational challenge is to identify the original materials (reverse engineering). Mass spectrometry is used in e.g.: (a) Proteomics to identify proteins in biological samples (e.g. blood), (b) Clinical applications to identify proteins related to cancer or other diseases, (c) Pharmaceutical analysis to determine the effects of new drugs, (d) Environmental contamination analysis to ensure that the air, drinking water, soils, and food are safe to consume and does not contain pollution, heavy metals, hormone, pesticides, and herbicides, (e) Forensic analysis to trace of evidence in arson investigation, drug abuse, and (f) Metabolomics to identify small molecules used by bacteria for communication in microbiome, etc.

There are several research projects available on BSc, Msc, PhD, and postdoc levels, which focus on development of deep and machine learning methods to identify mass spectrometry data. You will obtain a good understanding of the aspects and the challenges of computational mass spectrometry and deep learning with non-human readable data.

All projects are conducted in English, so it’s a good opportunity to improve your communication & presentation skills in English. All projects involve programming.

2021-present Laboratory Head, Laboratory on AI for Computational Biology, HSE University, Moscow, Russia,

2015-present Assistant Professor (Docent), HSE University, Moscow, Russia

2013-2015 Postdoctoral Fellow, Bill Noble's Lab University of Washington, Seattle WA, USA

2009-2013 Postdoctoral Fellow, Bioinformatics Group, International Centre of Genetic Engineering and Biotechnology (ICGEB), Trieste, Italy

2008-2009 Research Fellow, Division of Imaging and Applied Mathematics, CDRH, U.S. Food and Drug Administration (U.S. FDA), Silver Spring MD, USA. Joint affiliation with Department of Biology, University of Maryland Baltimore County (UMBC), Catonsville MD, USA

2004-2008 Ph.D. Student, University of Szeged, Hungary

2000-2004 Undergraduate Research Assistant, Research Group on Artificial Intelligence, Hungarian Academy of Sciences, Szeged, Hungary