Computational Math in Industry and Beyond (CME 500) - Winter

This session will take place online.

Covid19 reset the travel and tourism industry. Historical data is useless now and so are models built using historical data. So, how can data science help?

Arun is the Chief Data Scientist at BCG GAMMA, focused on building and deploying AI at Scale solutions. BCG GAMMA is BCG’s global team dedicated to applying artificial intelligence and advanced analytics to critical business problems at leading companies and organizations. The team includes more than 1500 data scientists, data engineers, machine learning engineers, software developers and UI/UX experts. Arun is a data scientist by training: PhD from Boston University in Computational Neuroscience (focusing on learning models -- what is now deep learning) and Masters from Northeastern University in Electrical and Computer Engineering. Arun was an adjunct faculty at Northwestern University for machine learning. He has several patents in deep learning and a few in the pipeline using reinforcement learning for industry use cases. Arun also leads the scientific review team at GAMMA – pushing innovation and collaboration with academia.

This session will take place online.

Geospatial data presents unique challenges from a computer-vision perspective. This talk will dive into some of the approaches taken to address these challenges and develop modeling approaches specifically tailored towards geospatial data.

Caroline is a deep learning scientist with a background in computer vision, currently living in Salt Lake City, UT. She works as a researcher on the computer vision team at the Palo Alto-based company Orbital Insight, where she designs novel architectures for use in the geospatial domain. She holds a BS in Biochemistry From UCLA, where she also engaged in vision-based neuroimaging research. In 2020, she earned an MS from Stanford University in Computational and Mathematical Engineering under the Imaging Science track. Prior to joining Orbital Insight, she was a Senior Data Scientist at Verisk Analytics, where she developed computer vision models for use in the insurance claims industry. Caroline enjoys playing ultimate frisbee, rock climbing, backpacking, and long-distance running. She is an avid Denver Nuggets fan and her newfound COVID-era hobbies include graphic design and Kaggling. In her spare time, you can also find her catching up on the newest computer vision papers and designing computer vision-based sports analytic systems.

This session will take place online.

PayPal believes that convenient, efficient, and affordable access to the financial system is a fundamental human right, not just a privilege for the affluent.  For over two decades, PayPal has leveraged technology to streamline digital payments and commerce; in 2020 the company processed over 15 billion transactions totaling nearly $1 trillion.

Last year, we created a new unit dedicated to harnessing blockchain technology and digital currencies to further streamline and democratize the financial services ecosystem.  Bringing PayPal’s standard of security, compliance, and ease-of-use to this new space has required us to double-down on our capabilities for risk management, transaction monitoring, and operations as well as to develop new models and techniques specific to digital currencies.

In this talk, Edwin will discuss how PayPal leverages data science and computational engineering to reduce fraud, foster greater inclusion, and protect customers as we work to bring the promise of digital assets to PayPal’s 400 million+ customers worldwide.

Edwin Aoki is a PayPal Technology Fellow and the CTO of the company’s Blockchain, Crypto, and Digital Currencies (BCDC) unit, where he is responsible for driving sustained growth and innovation in digital asset technology. Previously, he spent nearly a decade as the company’s Chief Architect, tasked with defining the company’s long-term technical roadmap and architecture and advancing the technologies and technologists that have enabled PayPal to become a fintech leader.

Before joining PayPal in 2010, Edwin was at Netscape and AOL for more than 13 years as Technology Fellow and Chief Architect, overseeing the architecture and technology strategy for many of AOL’s consumer facing products, including instant messaging, mail, mobile, enterprise and developer programs. Edwin's earlier roles include positions with Intuit, GO Corporation, and Apple Computer. 

Edwin graduated from Harvard College with a degree in Sociology and Computer Science; is a published author; and holds two dozen software patents. When he’s not at work, Edwin is involved with wildlife conservation and likes to spend time with big cats.

Automated speech recognition (ASR) systems are now used in a variety of applications to convert spoken language to text, from virtual assistants, to closed captioning, to hands-free computing. By analyzing a large corpus of sociolinguistic interviews with white and African American speakers, we demonstrate large racial disparities in the performance of popular commercial ASR systems developed by Amazon, Apple, Google, IBM, and Microsoft. Our results point to hurdles faced by African Americans in using increasingly widespread tools driven by speech recognition technology. More generally, our work illustrates the need to audit emerging machine-learning systems to ensure they are broadly inclusive. See more at fairspeech.stanford.edu.

Allison Koenecke is a postdoc at Microsoft Research in the Machine Learning and Statistics group, and starting Summer 2022 will be an Assistant Professor of Information Science at Cornell University.  Her research primarily spans two domains: algorithmic fairness in online services, and causal inference in public health.  Previously, she received her PhD from Stanford's Institute for Computational & Mathematical Engineering, and her Bachelor's from MIT in Mathematics with Computer Science.

This session will take place online.

Over the last 6 years, P&G has built an Industry Leading Data Science and AI Engineering Organizations focused on Transforming the company with Algorithms.  Jeff will discuss P&G’s long-time leadership in Analytics, its recent Data Science driven transformations, and what it means to have a career in Data Science. 

Jeffrey Goldman is Vice President of Enterprise Data Science, serving as analyst to the CEO and leading P&G’s Global Data Science Organization.  In recognition of his sustained analytic contribution to P&G, he was inducted into P&G’s IT Distinguished Experts.  Before his current role, Jeff led Business Analytics for Global Markets and the Western European Analytics Organization and founded the Business Analytics group for China and Product Supply Analytics for Asia. 

Ashwin leads Target’s global Artificial Intelligence team responsible for products involving Demand Forecasting, Inventory Planning & Control, Price Optimization, Personalized Recommendations, Search, and Marketing Science. He is also an Adjunct Professor in Applied Mathematics (ICME) at Stanford University where along with research and teaching in Reinforcement Learning, he directs the Mathematical and Computational Finance program.
The common theme in his diverse career across the Finance and Retail industries has been to create or boost business profitability through advanced yet pragmatic Mathematics & Engineering. He invests heavily in recruiting and mentoring rare talents, fostering a vibrant team culture, focusing on the RIGHT business problems to solve, and meeting challenging goals through diligent prioritization.

This session will take place online.

AI is ready to transform every industry by delivering new insights and capabilities that were previously un-imaginable. To get there, it will take new levels of performance in both hardware and software. This talk will explore what is needed to take an innovative new AI solution to production – End-to-End from Data prep, Modelling, and finally Deployment. 

Wei Li is Vice President and General Manager of Artificial Intelligence and Analytics at Intel Corporation, where he is responsible for AI software engineering and hardware co-design. His team has improved AI performance by 10-100X through software acceleration, and is instrumental in Intel's multi-billion dollar AI revenue growth. He received his Ph.D. in Computer Science from Cornell University on supercomputers, and taught Advanced Compiling Techniques at Stanford University.

This session will take place online.

Robotic surgery is the use of technology to perform complex medical procedures with more precision, flexibility, and control. Robotic Assisted Surgical Devices (RASDs) have the potential to reduce operation time and anesthesia use, minimize complication risk, and improve patient recovery. With the rapid acceptance of robotic surgery and slow-paced adoption, there exists several computational challenges for further exploring the limitations in functionality. This session will cover how optimization algorithms, ML, and model-based design are being used within the medical workflow, to perform pre-operative surgical planning and improve RASD design.

Moiz is the Manager of Electrical and Computer Engineering at MathWorks, leading development in several areas including robotics and AI. Moiz received his BS and MS from Rutgers University and New York University, in Applied Engineering and Biomechanics, respectively. He earned his PhD at Columbia University in Mechanical Engineering, working on rehabilitation robotic platforms. During his studies, Moiz developed and patented a posture training robot for assisting in rehabilitation of children with cerebral palsy and adults with spinal cord injury. Immediately following, he was a postdoctoral fellow at Harvard Medical School and the Brigham and Women’s Hospital, where his research focused on developing surgical planning and optimization algorithms and minimally invasive robotic devices. Moiz has also served as a consultant for several industry leaders on robotics and design.

This session will take place online.

Enterprise data is often stored in different formats (e.g. structured, unstructured, semi-structured) and scattered across different departments (e.g. HR, IT, …). This can cause data quality issues (e.g. incomplete data, inconsistencies, duplicates, …), and lead to inefficiencies in data-driven processes. Unifying different data sources can significantly improve data quality and lead to more accurate decisions. In this talk, we discuss how Knowledge Graphs and Entity-Centric AI techniques can be used to unify disparate data sources and bring better insights into downstream enterprise use cases. 

Ines Chami is a recent Ph.D. graduate from ICME at Stanford, where she was advised by Prof. Christopher Ré. Her research was focused on learning representations (embeddings) for graph-structured data such as Knowledge Graphs. She is now developing Entity-Centric AI technologies for enterprise data at Numbers Station, a stealth-mode startup. 

Speakers at the Women in Data Science Conference

Register for the online conference

This session will take place online.