Canada CIFAR AI Chair
CIFAR Azrieli Global Scholar 2016-2018
Learning in Machines & Brains
Pan-Canadian AI Strategy
Alona Fyshe is a Fellow in CIFAR’s Learning in Machines & Brains program, a CIFAR Azrieli Global Scholar 2016-2018, and Canada CIFAR AI Chair at Amii. She is an assistant professor.at the University of Alberta.
Fyshe is a computer scientist by training, but her research straddles multiple areas including neuroscience, machine learning, and computational linguistics. Her lab explores how the human brain processes and represents language meaning (semantics) and how it combines words to make complex meaning (semantic composition). Traditionally, models of semantics have been based on large text datasets. These are useful to a point, but to build complete semantic models we need to incorporate experiential knowledge. One way to do this is to study semantics ‘in vivo,’ using brain imaging technology such as fMRI and EEG.
Currently, brain image datasets are very small, representing about an hour of acquisition time per participant. Fyshe proposes moving brain imaging to ‘the field’ – having people record their own brain images, using consumer-grade EEG systems deployed in the home. This will produce much larger datasets, allowing her to more fully explore the complexities of language.
Haoyan, X., Murphy, B., & Fyshe, A. (2016). BrainBench: A Brain-Image Test Suite for Distributional Semantic Models. SIGDAT Conference on Empirical Methods for Natural Language Processing, Austin, Texas.
Fyshe, A., Talukdar, P.P, Murphy, B., & Mitchell, T.M. (2014). Interpretable Semantic Vectors from a Joint Model of Brain- and Text-Based Meaning. Annual meeting of the Association for Computational Linguistics, Baltimore, Maryland.
Wehbe, L., Murphy, B., Talukdar, P., Fyshe, A., et al. (2014). Simultaneously Uncovering the patterns of brain regions involved in different story reading subprocesses. Plos One.
Fyshe, A., Murphy, B., Talukdar, P., & Mitchell, T. (2013). Documents and dependencies: an exploration of vector space models for semantic composition. In Proceedings of the Seventeenth Conference on Computational Natural Language Learning (pp. 84-93).
Sudre, G. Pomerleau, D., Palatucci, M., Wehbe, L., Fyshe A., et al. (2012). Tracking neural coding of perceptual and semantic features of concrete nouns. Neuroimage, 62(1), 451–63. DOI: 10.1016/j.neuroimage.2012.04.048
CIFAR is a registered charitable organization supported by the governments of Canada, Alberta and Quebec, as well as foundations, individuals, corporations and Canadian and international partner organizations.