Toluwa focuses on leveraging tensor algebra, which deals with computations on multidimensional data, to understand and represent graph algorithms and other irregular applications. Prior tensor algebra tools have a proven track record of decomposing hard tensor algebra problems--which also have many design choices to make---into smaller, manageable pieces. Extending tensor algebra notation provides a succinct, elegant, and precise way to describe graph algorithms, enabling researchers to systematically explore and compare different graph solutions for a given problem, and hopefully reaching more efficient solutions. Toluwa’s long-term vision is a new framework of thinking about graph problems and an ecosystem of various tools that enable the automated exploration of different algorithms and implementations in both software and hardware platforms.
Toluwanimi (Toluwa) Odemuyiwa, is a PhD Candidate at the University of California, Davis, where she is advised by Prof. John Owens. Prior to Davis, she completed a masters degree in electrical engineering at San Jose State University while concurrently working as a full-time design verification engineer at Microsemi Corporation. Mentored by Prof. Birsen Sirkeci, her MSEE thesis focused on using machine learning to classify wireless radio signals. She obtained her BASc (Engineering Science) from the University of Toronto in 2017 working on using electroencephalography signals as a modality for biometric verification, advised by Prof. Dimitrios Hatzinakos. Her current research interests generally lie in algorithmic, software, and hardware techniques for finding novel ways to handle sparse data and computations as the need for high performance computing continues to grow in the age of Big Data. She is particularly interested in the ways software and hardware can be co-designed to produce novel solutions.