• Explainable AI @ Capital One Center for Machine Learning

    *** This is an online event. Zoom link will be shared with registered attendees *** Agenda (All Central times):[masked]pm: Event kickoff & Meetup Intro[masked]pm: Main Talk (Includes 30mins for Q&A) What are the top challenges companies face when deploying AI solutions? Most experts argue that the limitations around explainability of AI models continue to be among the top reasons AI solutions do not see the light of day, even if they significantly improve accuracy. As such, Explainable AI is considered the next frontier in AI space. Moreover, we cannot have Responsible AI without Explainable AI. In this talk, Teuta Mercado – Director of Responsible will lay out the context around why Explainable AI is required for Responsible AI, followed by Bayan Bruss - Head of Applied Research and Senthil Kumar-Chief ML Scientist at the Capital One Center for Machine Learning reviewing Explainable AI space along with some of the cutting edge research they are leading. About the speakers: Bayan Bruss Bayan is the director and head of Applied Machine Learning research at Capital One’s Center for Machine Learning. His research experience spans many disciplines from risk modeling for critical infrastructure, computational psycholinguistics, cybersecurity, and representation learning. Most recently he has been focused on the application of graph machine learning to financial transactions. He has served on the program committee for several conferences and workshops (KDD ‘20, ICAIF ‘20, Usenix SCaiNet ‘19). Bayan is reachable at: https://www.linkedin.com/in/bayan-bruss/ Senthil Kumar Senthil is the Chief ML Scientist at Capital One’s Center for Machine Learning where he applies Machine Learning and AI to various business problems. Prior to joining Capital One, he was at Bell Laboratories where he developed new technologies and managed several successful products that have been licensed around the world. He has published over 30 papers and holds 6 patents. Most recently, he co-organized the 2019 KDD Workshop on Anomaly Detection in Finance, and the 2019 NeurIPS Workshop on Robust AI in Financial Services. Senthil is reachable at: https://www.linkedin.com/in/senthil-kumar-24b0419/ Teuta Mercado Teuta leads the Responsible AI program at Capital One which seeks to empower associates to ethically innovate machine learning. Teuta joined Capital One in 2016 as a Compliance Advisor to the Retail Bank, and has over 10 years of experience in Financial Services compliance. Teuta is reachable at: http://linkedin.com/in/teuta-bitici-mercado-b9058a4

  • Virtual: Business Process Optimization with Machine Learning!

    This is a joint event with Kaggle Days Meetups, Dallas. Kaggle Days Meetups are a series of events held all over the world, created by Kaggle and LogicAI, that aim to gather Kagglers and people interested in data science around one city. Agenda: 4:00 - 4:10 Open lobby chat between virtual participants 4:10 – 4:15 Introductions by Babar, Aamer, and Authman 4:15 – 5:00 Presentations and QA 5:00 – Closing remarks and SPECIAL announcement! Abstract: Running inference with machine learning models has become rather straightforward. But the process needed to get there successfully can be anything but simple. Technical and business complications enforce constraints, additional ad-hoc training and validation might be required, perhaps even thousands of models may need to be spun up, or you might even do something unconventional like using a label to predict itself! In this meetup, we demonstrate with a real-life use case that properly understanding a business issue and then translating it into a technically constrained problem are KEY to applying data science properly. During the presentation, Kaggle Master and long time data-science educator Laurae, along with his invitee, Sebastien Davoust will talk about how they approach operational efficiency with "next-day package delivery routing" optimization at La Poste Colissimo, a French prime postal delivery service. From understanding the problem to implementing a solution, you will discover and learn the most important ins and outs of solving business problems with applied data science. Go beyond model.fit! Speakers: Laurae (Damien Soukhavong) was awarded an MSc with honors in auditing, management accounting, and information systems from SKEMA Business School in Paris. He spent five years doing data science freelancing, corporate ML training, and content writing. He has since been working as an architect and data science consultant at Planeum for large accounts, such as La Poste, Givaudan, and Faurecia. https://www.linkedin.com/in/dsoukhavong Sebastien Davoust received his Master in electronics, electrical engineering, and automatic from the Paris-Saclay University (formerly Paris Faculty of Sciences of Orsay). He worked as a Unix and BI consultant at Thales from[masked], and as a SAP BI expert consultant from[masked]. Since then, he's been working as a business intelligence coordinator at La Poste Colissimo. https://www.linkedin.com/in/sébastien-davoust-6a376726/ About our Hosts: Dialexa is a Technology Research, Design, and Creation Firm. We build revolutionary technology products and user experiences that solve today's complex business challenges. Whether it's bringing car subscriptions to market or re-engineering physician engagement for healthcare systems, Dialexa drives true innovation, creates larger impact, and furthers our client's business objectives. To learn more about Dialexa, visit https://www.dialexa.com #BusinessUnderstanding #OperationsResearch #PredictiveModeling #MachineLearning

  • Implications of Artificial Intelligence for Cybersecurity

    *** This is an online event. Zoom link will be shared with registered attendees *** [masked]pm: Kickoff and Introduction[masked]pm: Main Talk and Q&A Ever wonder about the impact of AI on Cybersecurity? Join us to hear from a top authority on this subject, Dr. Frederick Chang, former Director of Research at the NSA (National Security Agency) who has also served as a cybersecurity expert witness on multiple occasions at hearings convened by the U.S. House of Representatives. With ever growing cybersecurity threats and challenges, it’s no surprise that the interest on the impact of AI on cybersecurity has boomed. The computing and communications technologies on which we have come to rely present serious security concerns; Cyberattacks have escalated in number, frequency, and impact, drawing increased attention to the vulnerabilities of cyber systems and the need to increase their security. In the face of this changing landscape, there is significant concern and interest among policymakers, security practitioners, technologists, researchers, and the public about the potential implications of AI and ML for cybersecurity. The National Academies of Sciences, Engineering, and Medicine convened an extensive workshop on the topic last year and published their detailed findings at: https://www.nap.edu/catalog/25488/implications-of-artificial-intelligence-for-cybersecurity-proceedings-of-a-workshop. In this talk Dr. Chang, who chaired the workshop, will summarize those valuable finings. The talk is suitable for both technical and business audience. About the Speaker: Frederick R. Chang is the Chair of the Computer Science Department in the Lyle School of Engineering at Southern Methodist University (SMU). He is also the Bobby B. Lyle Endowed Centennial Distinguished Chair in Cyber Security and Professor in the Department of Computer Science. He is the Founding Director of the Darwin Deason Institute for Cyber Security and is a Senior Fellow in the John Goodwin Tower Center for Public Policy and International Affairs in SMU’s Dedman College. Additionally, Chang’s career spans service in the private sector and in government including as the former Director of Research at the National Security Agency. Dr. Chang was elected as a member of the United States National Academy of Engineering in 2016. He is currently the Co-Chair of the Intelligence Community Studies Board of the National Academies of Sciences, Engineering and Medicine and he is also a member of the Army Research Laboratory Technical Assessment Board of the National Academies. He has served as a member of the Computer Science and Telecommunications Board of the National Academies and as a member of the Commission on Cybersecurity for the 44th Presidency. He is the lead inventor on two U.S. patents and has appeared before Congress as a cybersecurity expert witness on multiple occasions. Dr. Chang received his B.A. degree from the University of California, San Diego and his M.A. and Ph.D. degrees from the University of Oregon. He has also completed the Program for Senior Executives at the Sloan School of Management at the Massachusetts Institute of Technology. He has been awarded the National Security Agency Director’s Distinguished Service Medal.

  • Introduction to Generative Adversarial Networks (GANs)

    Agenda:[masked]pm: Networking & food[masked]pm: Introduction to GANs and its applications by Taylor Brown Deep fakes are popular in news media. GAN or generative adversarial network is the technology behind these synthetic images. GAN is a machine learning (ML) model in which two neural networks compete with each other to become more accurate in their predictions. GANs are the most popular kind of deep learning models at this time. In this talk you will learn how generative models work and their applications for image manipulation and other business problems. About the Speaker: Taylor Brown is a principal data scientist with CoreLogic. He has worked on several machine learning research and commercial projects for the insurance and real estate markets. His interests include computer vision, deep learning and NLP. Linkedin Profle: https://www.linkedin.com/in/taylor-brown-ds/ More info: https://pathmind.com/wiki/generative-adversarial-network-gan

  • Graph Data Science - Improve predictions using connections in data

    Location visible to members

    [masked]: Networking & Food[masked]: Main Talk - Amy Hodler, Neo4j Graph Data Science - Improve predictions using connections in data The world is naturally connected and relationships are the strongest predictors of behavior. In this talk, you’ll learn how graphs (mathematical representations of connected data) can be used to improve predictions by taking advantage of relationships and network structures. We’ll walk through the steps of Graph Data Science employed alongside machine learning and AI systems including knowledge graphs, graph analytics and graph feature engineering. We’ll also review a link prediction workflow using graphs to increase ML model accuracy. Amy Hodler is a network science enthusiast and program director for AI and graph analytics at Neo4j. Amy is the co-author of the O’Reilly book, Graph Algorithms: Practical Examples in Apache Spark and Neo4j. She tweets @amyhodler Get the book from: https://neo4j.com/blog/new-oreilly-book-graph-algorithms-spark-neo4j/

  • Deep Learning and the Analysis of Time Series Data

    Agenda:[masked]pm: Networking & food[masked]pm: Main talk (Speaker: Dr. Bivin Sadler, Associate professor SMU MSDS Program) We are all familiar with the 4 Vs of Big Data: Velocity, Veracity, Volume and Variety. With respect to the fourth ‘V’, ‘Variety’, various unstructured data types such as text, image and video data have gained quite a bit of attention lately and continue to gain momentum. However, there is a type of structured data that has maintained its intrigue and importance in both business and academia … Time Series Data. While it is not hard to find applications of Deep Learning methods to text, image and video data, they have also shown to have great promise in the analysis of data collected over time. In this session, we will investigate the theory, implementation and performance of recurrent neural networks (RNNs), long short-term memory (LSTM), and gated recurrent units (GRUs) in the context of forecasting and predictive analytics. In addition, we will compare these deep learning methods with the more traditional but widely used ARMA and ARIMA type models. Join us for a fun and informative discussion on a topic that is showing great promise in an area critical to science, business and industry: Deep Learning and the Analysis of Time Series Data! About the Speaker: Originally from Dallas Texas, Dr. Bivin Sadler finished a BS in mathematics magna cum laude from Texas Tech University before beginning his professional career in Scottsdale, Arizona, at Motorola. He worked as a statistician and software engineer for 2.5 years, working primarily on a companywide tool to predict when software projects could be released with optimal statistical properties (Six Sigma). Upon completion of the project, he moved to San Diego, and while playing professional beach volleyball for two years, finished a master’s degree in applied math at San Diego State University. He then moved back to Dallas to earn a PhD in statistics from SMU and finished his degree in 2014 after winning the Walsh Award for the top score on the qualifying exam taken after the third year of coursework. Dr. Sadler was hired as part of the faculty at SMU after graduation and began a dual appointment teaching both undergraduate and graduate classes in the statistics department and online with the recently formed Master of Science in Data Science (MSDS) program. Academically, he has presented his work in item response theory at various conferences and is currently working on several domestic and international consulting projects. He became a full-time member of the MSDS faculty in August 2018 and, in addition to consulting projects and teaching, actively contributes towards developing new courses and enhancing existing ones at the SMU MSDS program.

  • Building LinkedIn for Things - AI and ML on the Edge

    [masked]: Networking & Food[masked]: Main Talk - Simon Crosby, CTO, Swim.AI Building LinkedIn of Things We have a special guest from California-based Swim.ai. Dr. Simon Crosby will talk about his vision on building a DataFabric for the distributed enterprise by converting streaming data from the edge to insights using ML. https://www.swim.ai/ About the speaker: Dr. Simon Crosby is the CTO at SWIM.AI, an edge-based software firm that executes real-time analytics and machine learning for IoT. Crosby brings an established record of technology industry success. Prior to SWIM.AI, he was co-founder and CTO of Bromium, a security technology company. At Bromium, Crosby built a highly secure virtualized system to protect applications. Earlier on he was the co-founder and CTO of XenSource, before its acquisition by Citrix, and later served as the CTO of the Virtualization and Management Division at Citrix. Previous to Citrix, Crosby was a principal engineer at Intel and also the founder of CPlane, a network-optimization software vendor. Dr. Crosby has been a tenured faculty member at the University of Cambridge.

  • AI/ML Use Cases in Telecom


    Agenda:[masked]pm: Networking & food[masked]pm: Main talk (Speaker: Swamy Vasudevan, VP of Digital Services at Ericsson) Telecom industry is embracing AI around the world. More than half of service providers plan to adopt it in their networks by 2020. The vision is to automate the managing of the networks as much as possible. Self-operating, self-optimizing and self-healing network is the way to go. Successful AI use cases include feature optimization in networks and failure predictions in telco sites. How can we best utilize AI to manage the machines ? About the Speaker: Swamy Vasudevan is the Vice president of Digital services responsible for sales, strategy and business development. Areas of focus include IT-Network Convergence, Virtualization, Machine Learning and Blockchain. He has around 25 years of industry experience both on business and technology. He has been instrumental in building new businesses in the organizations he worked for. Swamy is an industry thought leader in network communications & operations and has spoken at numerous industry events such as Openstack, NFV World Congress, Big telecom, TM Forum Management World, CTIA Wireless, Distributech, UTC, ESRI UC and others on topics such as Cloud Management, Smart Grid Communications, and GIS in communications network. He has been featured on February, 2016 cover of Vanilla Plus Telecoms IT journal; interviewed about virtualization solutions for the communications industry. Published feature in CIO magazine on SDN/NFV cloud management. Swamy holds a Bachelor of Engineering in computer science from PSG college of Technology and Master of Business Administration from Rutgers University.

  • Intro to Applied Reinforcement Learning

    2200 Commerce St

    [masked]pm: Networking & Food[masked]pm: Talk, Demo and Q&A Presenter: Dialexa -- https://www.dialexa.com Agenda: Deep reinforcement learning is exploding in the AI space thanks to advancements driven by projects like mastering Go, controlling robotic mechanics, and optimizing revenue management. Data-driven organizations are eager to extract the value from this upcoming form of machine learning but can struggle to frame a problem appropriately or build a game plan to execute on. This talk will walk you through the basics of reinforcement learning and how it differs from traditional machine learning, where and how you can apply it in your organization, and a demonstration of an agent learning in action. The talk will be lead by three senior machine learning engineers at Dialexa. Each has experience designing and implementing reinforcement learning environments and algorithms in previous projects. Rowdy Howell graduated from SMU in 2015 and spearheads the growing data science practice at Dialexa. Mallory Hightower is an SMU graduate student entering the applied machine learning space. Jonathan Rebello is a Georgia Tech and University of Texas graduate with five years of experience in preventative maintenance and risk optimization in the oil and gas industry. Finally, Uthman Apatira’s online data science courses have been taken by over 100,000 students internationally.