AI in Finance & RegTech Virtual Summit AI in Finance & RegTech Virtual Summit schedule

AI Beyond Pattern Recognition: Decision Making Systems

While machine learning and artificial intelligence technologies are now advanced enough to outperform humans in variety of tasks, how we make decisions with models varies by practitioner. Reinforcement learning is promising but it is limited to adversarial settings; or, in vernacular, situations where decisions directly impact the environment. Without figuring out how AI systems can make good decisions in environments they cannot influence, we may forever be stuck in a limbo of pattern recognition, prediction, and analytics. What if we can develop a “theory” of AI decision making? Can we view different decision making situations as a set of engineering systems? Can we define key components of an AI decision maker? Answering such questions would enable us to design AI systems in a modular fashion much like how we design many industrial goods like cars. We may even be able to develop industry standards and manuals on how to design AI decision makers. Using actual use cases and other potential real-world applications in both financial and non-financial settings as examples, a systems view on decision making AI systems is proposed. Furthermore, ways to design and build such systems are explored.

Key Takeaways: 1) AI systems are far more valuable making decisions than making simple predictions and pattern recognitions

2) We need a “theory of design” for AI decision making systems: For AI to become a trusted part of decision making both in and out of industry, we need to understand and generalize components of AI systems and define what it means to be “robust”.

3) This starts with looking at actual use cases, identifying similarities, and quantifying key parameters – so we may use standard design techniques to design AI systems.

Kevin is a data scientist with strong interest in an interdisciplinary approach that combines artificial intelligence, operations research, systems engineering, economics, and quantitative finance. He is a member of Nasdaq’s Machine Intelligence Lab, a team dedicated to using AI to improve capital markets. At Nasdaq, he has worked on projects that cover topics such as alternative data, capital market operations, financial surveillance, and portfolio management. His most recent interest is in developing a procedure for designing robust and fail-proof decision-making AI systems. Kevin holds a Bachelor’s degree in Computer Science from Washington University in St. Louis.

PRESENTATION: Reduce Compliance Risks with Trustworthy AI

Increasing complexity in the regulatory environment, rates of regulatory change and need for accountability are driving new compliance risks for financial services companies. Trustworthy AI reduces compliance risks while balancing human control and oversight with accountability. AI can automatically identify relevant regulatory changes and predict the impact to the organization (e.g., business units, policies, controls, products/services, contracts). Regulatory changes can be routed to impacted business units with compliance risk indicators/ratings and impact predictions. The AI solution can predict and automatically take actions necessary to maintain compliance, which can be accepted or overridden by the business unit.

ROUNDTABLE: Defining a Governance Framework Covering the Appropriate Level of Automation/Human Control

Points of Discussion: Defining compliance and regulatory risks Identifying and rating regulatory/organizational changes Defining compliance risk treatment strategies Including user feedback to improve the AI-enabled compliance management

Brian is responsible for the strategy and management of Omina Technologies US. He has a B.S. Mech. Eng. and a J.D. Brian has spent the last 25 years working with technology companies and new technologies in capacities ranging from legal advisor to executive to investor. While working in private law practice, Brian represented technology organizations in intellectual property, regulatory and litigation matters. Brian also has advised technology, financial services, energy/utilities and healthcare companies regarding corporate risk and compliance. Prior to joining Omina, Brian worked for C2C with responsibilities including internal legal advisory, corporate strategy and software development, as well as client management on numerous projects covering diverse risk/compliance matters such as information/cybersecurity and data privacy.

Insights on AI in Finances

Manuela M. Veloso recently joined J.P.Morgan Chase to create and head an Artificial Intelligence (AI) Research Center. Veloso is on leave from Carnegie Mellon University (CMU) where she is Herbert A. Simon University Professor in the School of Computer Science, and where she was the Head of the Machine Learning Department until June 2018. She researches in AI, Robotics, and Machine Learning. Veloso is AAAI Fellow, ACM Fellow, AAAS Fellow, and IEEE Fellow, Einstein Chair Professor of the Chinese National Academy of Science, the co-founder and past President of RoboCup, and past President of AAAI. Veloso and her students research a variety of autonomous robots, including mobile service robots and soccer robots. See www.cs.cmu.edu/~mmv for further information, including publications.

Customer centricity is the core of every insurance strategy, but how can AI and specifically NLP help carriers put customers first when it comes to claims? The challenges around speed and quality remain, but technology proved to bring tangible results.

In this session we will cover:

• Top challenges when it comes to automating the claims process? • How to automate claims processes? • The main KPIs to watch when it comes to claims automation using NLP? • Best practices and a few real-world case studies

Pamela Negosanti is the Global Head of Insurance at expert.ai. She specializes in helping companies in the insurance space, such as carriers, brokers, reinsurers, and TPAs execute business transformation through artificial intelligence. Pamela has spent most of her career in the technology sector where she has established expertise in artificial intelligence, cognitive computing, intelligent automation, natural language understanding and semantics, to name a few. She owns a degree in translating and interpreting with a specialization in computational linguistics, and speaks fluent Italian, English, French and German. Fueled by change and curiosity, she is firm believer in the power of knowledge share and communities.

Yanshuai Cao is a Senior Research Team Lead at Borealis AI, the AI R&D Institute of RBC. He has done a range of works in AI, from data science consulting to research in core machine learning, computer vision, and natural language processing. Recently Yanshuai and his team have focused on (text-to-SQL) semantic parsing, a problem central to natural language database interface and the next generation conversational AI. Yanshuai received his Ph.D. from the University of Toronto in Computer Science.

Conversational AI Solutions for Combatting Insurance Fraud

Speech and natural language technology have advanced at a rapid pace in recent years. This advance, a facet of the industry 4.0 era, has been driven in part by GPU hardware and the deep learning frameworks that use them, and by the adoption of open-source ​software by the academic and commercial AI community alike. These developments have markedly impacted the way in which humans communicate with computers and are currently driving numerous commercial products that rely on speech, natural language processing and natural language understanding, loosely termed Conversational AI. This talk will introduce the machine learning and deep learning approaches, which enable Conversational AI, and present a real-world case study in the insurance domain that exploits speech and language to detect deception and tackle fraud.

Dr Julie Wall is a Reader in Computer Science, Director of Impact and Innovation for the School of Architecture, Computing and Engineering and leads the Intelligent Systems Research Group at the University of East London. Her current research focuses on developing machine learning and deep learning approaches for speech enhancement, natural language processing and natural language understanding and she maintains collaborative R&D links with industry. This has led to the successful acceptance of two Innovate UK grants with a combined total value of £2,273,177. Since starting her PhD in 2006, Julie has been exploring the overarching research area of designing intelligent systems for processing and modelling temporal data. This primarily involves investigating the architectures and learning algorithms of neural networks for a variety of data sources.

https://www.uel.ac.uk/research/intelligent-systems

Dr. Ehsan Amjadian earned his Ph.D. in Deep Learning & Natural Language Processing from Carleton University, Canada. He is published in a variety of additional Artificial Intelligence and Computer Science domains including Cybersecurity, Recommender Engines, Information Extraction, and Computer Vision. He is currently the Director of AI & Technology at Royal Bank of Canada (RBC), where he has led numerous advanced AI products from ideation to production and has filed patents in the areas of Data Protection and Finance & Climate. He has also led various open source artificial intelligence initiatives as well as a multitude of research teams. He is a member of IEEE as well as ACL.

Using Deep Learning with Word Embeddings to improve Customer Satisfaction

Understanding customer satisfaction in retail banking requires exploring and comprehending multiple sources of feedback, such as emails, social networks reviews, web feedback, bot interactions, as well as speech-to-text transcripts collected from call centers. Since such a vast amount of textual data can be difficult to leverage with traditional text mining techniques, deep learning and word embeddings can be used to automatically classify and label feedback, and then deeply analyze and understand their content. In this communication we explain how we leverage all those AI techniques to get an in-depth understanding of the opinions and needs of National Bank’s retail customers. We also show how we improve the performance levels of those AI tools using in-house algorithms and data resources to improve the overall capacity of natural language understanding.

Key Takeaways: • Industrial applications • State of the art classification • Understanding of DL embedding limits

Eric Charton holds a Master in machine learning applied to voice recognition, and a Ph.D. in machine learning applied to Information extraction and natural language generation. He worked as scientist and research project coordinator in academic context in Europe (University of Avignon) and North America (CRIM, École Polytechnique de Montréal) before becoming head of search engine research and development at Yellow Pages Canada. Since March 2018, he is Senior AI Director at National Bank of Canada.

PRESENTATION: Speech Recognition: Comprehensive Insights in Real-World AI for Business Leaders

Learn how to streamline your workload, boost efficiency and automate processes with Dragon Speech Recognition. This live short tutorial will show you how you can overcome challenges such as documentation, collaboration across teams in real time and how to get key business insights that will help you perform better as a business and improve your services to customers.

ROUNDTABLE: AI & Speech Recognition in Financial Services

Economies are responding to many challenges (e.g. Brexit and the Covid-19 pandemic). Businesses need to find ways to remain both competitive and sustainable. As markets and profitability wax and wane under these influences, the financial services sector faces real and present pressures. How can we advance speech recognition in this time, and consequently build better products & provide better services?

Vladimir Teodosiev is a Sales Manager at Nuance delivering value to high-profile private and public organisations. Commercially savvy and with a strong interest in technology, he has extensive experience in creating and delivering value to high-profile private and public organisations. Highly customer-focused and results-driven, he supports businesses in developing best practices that facilitate efficient document creation.

Developing Early Warning System to Identify Relevant Events in Unstructured Data

Swiss Re is a leading player in the global reinsurance sector. Its role is to anticipate, understand and price risk in order to help insurers manage their risks and absorb some of their biggest losses. As one way to stay ahead of the curve and provide thought leadership to its clients, Swiss Re is developing an early warning expert community platform based around big data and natural language processing. The platform is intended to work on the front lines, to detect events that have the potential to change our view on risk drivers and to help us make business decisions in shorter timescales.

Key Takeaways: 1) The traditional methods to identify relevant events become unreliable when information volume rapidly increases; 2) Uncoordinated views pose a challenge in taking proactive and strategic actions to manage risks; 3) Early warning expert community platform leverages new data techniques to identify relevant signals and helps integrating experts in a more joined up process.

Nataliya Le Vine is a data scientist at Advanced Analytics Center of Excellence at Swiss Re, bringing machine learning and AI to drive the technology transformation in insurance. Over the last decade, she worked in academia, tech and insurance industries both in EMEA and Americas with a core expertise in predictive modeling and machine learning.

Annie leads the global product team of Swiss Re’s Risk Data Service platform along with various client risk management services for L&H insurers.

A recent example of Annie’s work is with the Swiss Re Risk Resilience Center Center. The ambition is to join the forces of academia, insurers, and data partners to solve Covid-related problems that are most relevant to the insurance industry and beyond.

In a past life, she has over a decade experience in Reserving, Inforce Management and Experience Studies. Annie is an eternal optimist and brings unparalleled energy and passion into her work.

PRESENTATION: Reduce Compliance Risks with Trustworthy AI

Increasing complexity in the regulatory environment, rates of regulatory change and need for accountability are driving new compliance risks for financial services companies. Trustworthy AI reduces compliance risks while balancing human control and oversight with accountability. AI can automatically identify relevant regulatory changes and predict the impact to the organization (e.g., business units, policies, controls, products/services, contracts). Regulatory changes can be routed to impacted business units with compliance risk indicators/ratings and impact predictions. The AI solution can predict and automatically take actions necessary to maintain compliance, which can be accepted or overridden by the business unit.

ROUNDTABLE: Defining a Governance Framework Covering the Appropriate Level of Automation/Human Control

Points of Discussion: Defining compliance and regulatory risks Identifying and rating regulatory/organizational changes Defining compliance risk treatment strategies Including user feedback to improve the AI-enabled compliance management

Brian is responsible for the strategy and management of Omina Technologies US. He has a B.S. Mech. Eng. and a J.D. Brian has spent the last 25 years working with technology companies and new technologies in capacities ranging from legal advisor to executive to investor. While working in private law practice, Brian represented technology organizations in intellectual property, regulatory and litigation matters. Brian also has advised technology, financial services, energy/utilities and healthcare companies regarding corporate risk and compliance. Prior to joining Omina, Brian worked for C2C with responsibilities including internal legal advisory, corporate strategy and software development, as well as client management on numerous projects covering diverse risk/compliance matters such as information/cybersecurity and data privacy.

PRESENTATION: Power Up Your Visual AI with Synthetic Data

Computer Vision is rapidly changing the retail landscape with respect to both the customer experience and in-store day to day logistics like inventory monitoring, brand logo detection, shopper behavior analysis , autonomous checkout. Traditional methods of training models with real world data is becoming a big bottleneck to faster deployment of these vision models. Learn how machine learning and computer vision engineers are using Unity to get faster, cheaper and more unbiased access to high quality synthetic training data and accelerating model deployment.

Key Takeaways: 1) Computer vision is becoming essential in retail with applications ranging from planogram verification to inventory monitoring to cashier-less checkout. 2) Labeled data is critical to computer vision but the traditional approach of using real-world training data is expensive, time-consuming, and often insufficient for training a production-level system. In contrast, synthetic datasets are less expensive, faster to produce, perfectly labeled, and tailored with the end application in mind. 3) Unity has technology to produce synthetic datasets with structured environments and randomizations that lead to robust model performance. This presentation shows samples from a Unity retail-oriented dataset that you can download.

ROUNDTABLE: Solving Retail Challenges with Computer Vision

Join Unity Computer Vision Experts and peers to discuss the rapidly growing field of computer vision and how it is impacting the retail world and some of the challenges associated with deploying computer vision in Retail. This is a freeform session where you can come to the table with your questions and we will have an engaging and interactive conversation around those topics. You can also use this time to talk with the Unity team about using synthetic data for training computer vision models and dig deeper into customer stories and proof points around synthetic data

James has 14 years of experience building and applying simulation and artificial intelligence technologies. He started his career in the simulation brand at Dassault Systèmes, where he worked on mechanical simulation solutions for automotive and aerospace customers. He spent several years managing the delivery of natural language systems for Alexa at Amazon. He has worked as a product manager in the AI organization at Unity since 2019 focusing on Unity Simulation and Unity’s solutions for computer vision and is excited about the next frontiers in AI.

AI Beyond Pattern Recognition: Decision Making Systems

While machine learning and artificial intelligence technologies are now advanced enough to outperform humans in variety of tasks, how we make decisions with models varies by practitioner. Reinforcement learning is promising but it is limited to adversarial settings; or, in vernacular, situations where decisions directly impact the environment. Without figuring out how AI systems can make good decisions in environments they cannot influence, we may forever be stuck in a limbo of pattern recognition, prediction, and analytics. What if we can develop a “theory” of AI decision making? Can we view different decision making situations as a set of engineering systems? Can we define key components of an AI decision maker? Answering such questions would enable us to design AI systems in a modular fashion much like how we design many industrial goods like cars. We may even be able to develop industry standards and manuals on how to design AI decision makers. Using actual use cases and other potential real-world applications in both financial and non-financial settings as examples, a systems view on decision making AI systems is proposed. Furthermore, ways to design and build such systems are explored.

Key Takeaways: 1) AI systems are far more valuable making decisions than making simple predictions and pattern recognitions

2) We need a “theory of design” for AI decision making systems: For AI to become a trusted part of decision making both in and out of industry, we need to understand and generalize components of AI systems and define what it means to be “robust”.

3) This starts with looking at actual use cases, identifying similarities, and quantifying key parameters – so we may use standard design techniques to design AI systems.

Kevin is a data scientist with strong interest in an interdisciplinary approach that combines artificial intelligence, operations research, systems engineering, economics, and quantitative finance. He is a member of Nasdaq’s Machine Intelligence Lab, a team dedicated to using AI to improve capital markets. At Nasdaq, he has worked on projects that cover topics such as alternative data, capital market operations, financial surveillance, and portfolio management. His most recent interest is in developing a procedure for designing robust and fail-proof decision-making AI systems. Kevin holds a Bachelor’s degree in Computer Science from Washington University in St. Louis.

Machine Learning Engineer at ING. Currently, I am using NLP to extract key information from documents. I have also worked with credit risk analysis using machine learning models for SME customers. Before joined ING, I worked in Accenture and IBＭ with AI projects with anti-money laundering and financial assistant topics. Coming from an academic AI background, I always keen on applying the state of art models to real-life financial problems. Nevertheless, my passion does not limit in R&D, but also bring the models in production and benefit users.

Data-driven and AI solutions have long been heralded as the new game changer in almost all industries. Finance is no exception. With the ever growing number of data sources, interaction points, and patterns of product usage, data science and AI must be an integral part of any business area engaged with risk management and AML. One such seamless integration requires the synchronisation of many functions and introduction of new processes and responsibilities. In this talk we will gloss over the journey at Swedbank and look ahead to a future with no shortage of exciting challenges and opportunities

Mehrdad Mamaghani is the head of Department for Data Science at Swedbank. The Department is the bank’s Centre of Excellence for Analytics & AI and focuses mainly on financial crime and cybersecurity.

Mehrdad has a PhD in applied mathematical statistics and has spent the last three years building data science capabilities in the bank within a wide variety of applications and business domains.

Chris Merz has been with Mastercard for 18 years. He leads several teams of data scientists that provide an AI layer to such security and decision products as Decision Intelligence, Safety Net, Smart Authentication, AI Express, Stand In, AML, and other R&D efforts. Chris has also spent time in Mastercard Labs, Loyalty Solutions, and Advisors, applying machine learning techniques to engage cardholders. Before joining Mastercard, he worked at a startup for online personalization and performed machine learning research at the NASA COE for AI, and the McDonnell Douglas Research Lab. He has a PhD in Machine Learning from the University of California, a Master of Science in Computer Science from Missouri University of Science and Technology, and a Bachelor of Science in Computer Science from the University of Missouri, St. Louis.