2020 Morning Technical Talks
Data Scientist, Dell Technologies IoT Division, VMware
Fernanda Campello is a data scientist in the Dell Technologies Edge/IoT Division at VMware. As a data scientist, Fernanda works on developing analytics methodologies and use cases to gain insight and unlock the value from data, going from business and analytics problem framing, to model development and deployment. Before joining the Dell Technologies Edge/IoT Division, Fernanda was a senior data science consultant at Dell EMC, developing analytics projects in several industries, including healthcare, utilities, finance, insurance, and biotech. Fernanda holds a PhD in Operations and Information Systems from the University of Alberta in Canada, a Master’s in Industrial Engineering and a Bachelor’s Degree in Electrical Engineering from the Universidade Federal de Pernambuco, in Brazil. Fernanda has also worked in research and development of new analytical models, having published academic papers in the area of operations and information systems.
Talk Title: Explainable ML: understanding and trust in automated decisions
Abstract: As Machine Learning (ML) models continue to be used in a growing variety of applications, including support for potentially life-altering decision making in highly regulated industries such as banking, insurance, and healthcare, there is an accompanying growing need to improve understanding of how these automated decisions are being made. Explainable ML methods aim at addressing this need by increasing the ability to explain or to present the prediction process of complex models in understandable terms to human stakeholders. This talk will include an overview of the growing need for explainability in ML applications, as well as a discussion of factors impacting explainability, different methodologies available, and associated challenges.
Dr. Erin Solovey
Assistant Professor, Computer Science, WPI
Erin T. Solovey is an Assistant Professor of Computer Science at WPI. Dr. Solovey’s research expertise is in emerging human-computer interaction modes and techniques, such as brain-computer interfaces, physiological computing, wearable computing, and reality-based interaction. She designs, builds and evaluates interactive systems that use machine learning approaches to adapt and support the user’s changing cognitive state and context. She also investigates improving information accessibility for deaf individuals. Her work has applications in areas such as education, transportation, medicine, creativity support, gaming, and complex decision making. Her research is supported by NSF and has received awards including the NSF/CRA Computing Innovation Fellowship and three ACM CHI Best Paper Award Honorable Mentions. She is the Deputy Editor of the International Journal of Human-Computer Studies and regularly serves on the program committee for the ACM CHI conference on Human Factors in Computing Systems. She received a bachelor’s degree in computer science from Harvard, and her Masters and Ph.D. in computer science from Tufts. Before joining the WPI faculty, she was a professor at Drexel University and a postdoctoral fellow in the Humans and Automation Lab at MIT.