Friday, November 25, 2022
2:00pm-3:00pm
Myhal Centre, Room 360
55 St George St.
CARTE Industry Speaker Seminar Series welcomes Nima Safaei, Senior Data Scientist at Scotiabank, for the second in-person seminar of 2022-23 academic year.
Registration: To register, please see here. Capacity is limited. Please register early to secure your spot.
Abstract:Correspondent Banking (CB) Network refers to a network of financial institutions providing cross-border payment services for customers through different channels such as SWIFT, Fedwire, etc. Through the CB network, banks and their customers can access financial services in different jurisdictions and provide cross-border payment services to their customers, supporting, among other things, international trade and financial inclusion. We employ the mathematical programming approach in conjunction with the graph theory to optimize a CB network. Optimizing the network requires decisions to be made to onboard, terminate or restrict the bank relationships to optimize the size and overall risk of the network. This study provides theoretical foundation to detect the components, the removal of which does not affect some key properties of the network such as connectivity and diameter. We find that the correspondent banking networks have a feature we call k-accessibility, which helps to drastically reduce the computational burden required for finding the above mentioned components.
Speaker Bio:Nima Safaei holds a Ph.D. in System and Industrial Engineering with a background in Applied Mathematics. He held a postdoctoral position at C-MORE Lab (Center for Maintenance Optimization & Reliability Engineering), University of Toronto, Canada, working on Machine Learning and Operations Research (ML/OR) projects in collaboration with various industries and service sectors. He was with Bombardier Aerospace with a focus on ML/OR methods for reliability/survival analysis and airline operations optimization. Nima is currently with Scotiabank, Toronto, Canada, as senior data scientist; focusing on ML/OR methods for various financial/market use cases including prediction, explain-ability, causality Inference, and early warning signal detections. He has more than 40 peer-reviewed articles and book chapters published in top-tier journals. He has also been invited to present his research findings in top ML conferences such as GRAPH+AI, NVIDIA GTC, TMLS, and ICML.
