Stefan C. Kremer

Professor

College of Computational, Mathematical and Physical Sciences, School of Computer Science

Google Scholar

Research Gate Profile

ORCID

Seeking academic or industry partnerships in the area(s) of

Machine Learning/Pattern Induction, Biology/Bioinformatics

Available positions for grads/undergrads/postdoctoral fellows

Limited positions are available to outstanding candidates with exceptional academic records and proven research experience

Research Areas

Spatio-Temporal Pattern Recognition
Recurrent Networks
Pattern Induction
Machine Learning
Deep-learning
Deep Belief Networks
Bioinformatics

Education and Employment Background

Dr. Kremer completed his Bachelor of Science in Computing and Information Science at the University of Guelph and received his PhD from the University of Alberta. He joined the School of Computer Science at the University of Guelph in 1997, and during this time Professor Kremer served as the Director of the nascent Bioinformatics program (2008-2011) followed by Director of the School of the School of Computer Science (2011-2016).

Research Themes

Professor Kremer is interested in the problem of learning to recognize, categorize, and generate structural patterns based on examples (this is called induction). Structural patterns refer to patterns in data that are not easily represented by a fixed-length sequence of numbers (i.e. a vector). Examples include English text, the shape of a molecule, a sequence of DNA, a radio signal, a system log, or a video. Professor Kremer’s research is focused on the following major themes:

Architectures, algorithms and approaches to recurrent neural networks for learning structured data. Professor Kremer researches methods to elucidate structural patterns in data. His work focusses on using recurrent networks for sequences, recursive networks for other structures. He develops novel architectures and algorithms and seeks out methods for applying them to datasets from a variety of domains.
Applications of sequence learning. Professor Kremer is working on applying structure learning to problems in environmental science, biology and bioinformatics. For example, he has been collaborating with a team of biologists and a philosopher of science that is studying Transposable Elements (TEs). TEs are small sequences of DNA that act as parasites on a genomic level, coopting a body’s own reproductive mechanisms to propagate in a manner like viruses. Professor Kremer is focusing on developing models and interpretations of TE activity from both evolutionary and ecological perspectives. Additionally, Dr. Kremer works on developing methods to identify and reidentify animals in camera trap images and videos for wildlife preservation and ecology.

Highlights

Genome Canada, Bioinformatics and Computational Biology Competition (collaborator): Extracting Signal from Noise: Big Biodiversity Analysis from High-Throughput Sequence Data, 2017-2018
President's Distinguished Professor Award, University of Guelph, 2001-2002.