Timezone: »

Challenges and Opportunities for AI in Financial Services: the Impact of Fairness, Explainability, Accuracy, and Privacy
Manuela Veloso · Nathan Kallus · Sameena Shah · sk Kumar · Isabelle Moulinier · Jiahao Chen · John Paisley

Fri Dec 07 05:00 AM -- 03:30 PM (PST) @ Room 511 CF

The adoption of artificial intelligence in the financial service industry, particularly the adoption of machine learning, presents challenges and opportunities. Challenges include algorithmic fairness, explainability, privacy, and requirements of a very high degree of accuracy. For example, there are ethical and regulatory needs to prove that models used for activities such as credit decisioning and lending are fair and unbiased, or that machine reliance doesn’t cause humans to miss critical pieces of data. For some use cases, the operating standards require nothing short of perfect accuracy.

Privacy issues around collection and use of consumer and proprietary data require high levels of scrutiny. Many machine learning models are deemed unusable if they are not supported by appropriate levels of explainability. Some challenges like entity resolution are exacerbated because of scale, highly nuanced data points and missing information. On top of these fundamental requirements, the financial industry is ripe with adversaries who purport fraud and other types of risks.

The aim of this workshop is to bring together researchers and practitioners to discuss challenges for AI in financial services, and the opportunities such challenges represent to the community. The workshop will consist of a series of sessions, including invited talks, panel discussions and short paper presentations, which will showcase ongoing research and novel algorithms.

Fri 5:35 a.m. - 5:50 a.m.
Opening Remarks (Talk)
Manuela Veloso · Isabelle Moulinier
Fri 5:50 a.m. - 6:10 a.m.
Invited Talk 1: Fairness and Causality with Missing Data (Invited Talk)
Madeleine Udell
Fri 6:10 a.m. - 6:30 a.m.
Invited Talk 2: Building Augmented Intelligence for a Global Credit Rating Agency (Invited Talk)
Sameena Shah
Fri 6:30 a.m. - 7:30 a.m.
Panel: Explainability, Fairness and Human Aspects in Financial Services (Panel Discussion)
Madeleine Udell · Jiahao Chen · Nitzan Mekel-Bobrov · Manuela Veloso · Jon Kleinberg · Andrea Freeman · Samik Chandarana · Jacob Sisk · Michael McBurnett
Fri 7:30 a.m. - 8:00 a.m.
Coffee Break and Socialization (Break)
Fri 8:00 a.m. - 8:20 a.m.
Invited Talk 3: Fairness in Allocation Problems (Invited Talk)
Michael Kearns
Fri 8:20 a.m. - 9:00 a.m.

11:20 - 11:28 In (Stochastic) Search of a Fairer Alife 11:28 - 11:36 Where's the Bias? Developing Effective Model Goveranance 11:36 - 11:44 Scalable Graph Learning for Anti-Money Laundering: A First Look 11:44 - 11:52 Fair Resource Allocation in a Volatile Marketplace 11:52 - 12:00 Algorithmic Confidence – A Key Criterion for XAI and FAT

Fri 9:00 a.m. - 10:30 a.m.
Lunch (Break)
Fri 10:30 a.m. - 10:50 a.m.
Invited Talk 4: When Algorithms Trade: Modeling AI in Financial Markets (Invited Talk)
Michael Wellman
Fri 10:50 a.m. - 11:10 a.m.
Invited Talk 5: ML-Based Evidence that High Frequency Trading Has Made the Market More Efficient (Invited Talk)
Tucker Balch
Fri 11:10 a.m. - 11:50 a.m.

02:10 - 02:18: Accurate, Data-Efficient Learning from Noisy, Choice-Based Labels for Inherent Risk Scoring 02:18 - 02:26: An AI-based, Multi-stage detection system of banking botnets 02:26 - 02:34: Robust Classification of Financial Risk 02:34 - 02:42: An Empirical Evaluation of Deep Sequential Models for Volatility Prediction 02:42 - 02:50: Computer-Assisted Fraud Detection, from Active Learning to Reward Maximization

Fri 11:50 a.m. - 12:00 p.m.
Announcement: FICO XAI Challenge Winners (Announcement)
Fri 12:00 p.m. - 12:30 p.m.
Coffee Break (Break)
Fri 12:30 p.m. - 12:50 p.m.
Invited Talk 6: Is it possible to have interpretable models for AI in Finance? (Invited Talk)
Cynthia Rudin
Fri 12:50 p.m. - 1:30 p.m.

15:50 - 15:58 Use of Machine Learning Techniques to Create a Credit Score Model for Prepaid Basic Services in East Africa. Case Study: Airtime Loans 15:58 - 16:06 Towards Global Explanations for Credit Risk Scoring 16:06 - 16:14 Interpretable Credit Application Predictions With Counterfactual Explanations 16:14 - 16:22 Interpretable Feature Selection Using Local Information for Credit Assessment 16:22 - 16:30 Towards Explainable Deep Learning for Credit Lending: A Case Study

Fri 1:30 p.m. - 2:40 p.m.
(Poster Session)
  1. Clustering and Learning from Imbalanced Data
  2. Deep Hedging: Hedging Derivatives Under Generic Market Frictions Using Reinforcement Learning
  3. Generating User-friendly Explanations for Loan Denials using GANs
  4. Practical Deep Reinforcement Learning Approach for Stock Trading
  5. Idiosyncrasies and challenges of data driven learning in electronic trading
  6. Machine learning-aided modeling of fixed income instruments
  7. An Interpretable Model with Globally Consistent Explanations for Credit Risk
  8. Continuous learning augmented investment decisions
  9. HELOC Applicant Risk Performance Evaluation by Topological Hierarchical Decomposition
  10. Looking Deeper into the Deep Learning Models: Attribution-based Explanations of TextCNN
  11. Matrix Regression and Its Applications in Cryptocurrency Trading
  12. Sensitivity based Neural Networks Explanations
  13. On the Need for Fairness in Financial Recommendation Engines
  14. Read the News, not the Books: Predicting Firms’ Financial Health
Ramya Malur Srinivasan · Miguel Perez · Yuanyuan Liu · Ben Wood · Dan Philps · Kyle Brown · Daniel Martin · Mykola Pechenizkiy · Luca Costabello · Rongguang Wang · Suproteem Sarkar · Sangwoong Yoon · Zhuoran Xiong · Enguerrand Horel · Drew Zhang · Ulf Johansson · Jonathan Kochems · Gregory Sidier · Prashant Reddy · Lana Cuthbertson · Yvonne Wambui · Christelle Marfaing · Galen Harrison · Irene Unceta Mendieta · Tom Kehler · Mark Weber · Li Ling · Ceena Modarres · Abhinav Dhall · Arash Nourian · David Byrd · Ajay Chander · Xiao-Yang Liu · Hongyang Yang · Shuang (Sophie) Zhai · Freddy Lecue · Sirui Yao · Rory McGrath · Artur Garcez · Vangelis Bacoyannis · Alexandre Garcia · Lukas Gonon · Mark Ibrahim · Melissa L
Fri 2:40 p.m. - 2:45 p.m.
Closing Remarks (Talk)

Author Information

Manuela Veloso (JPMorgan and Carnegie Mellon University)
Nathan Kallus (Cornell University)
Sameena Shah (S&P Global)

Sameena is a Managing Director and Director of Research at JP Morgan AI Research. Previously Sameena was a Managing Director and Head of Data Science at S&P Global Ratings. Prior to that she led all AI strategy, Research and development for Financial and News verticals of Thomson Reuters, and worked as a quant at a hedge fund. Sameena has won several best paper awards, has 40+ publications, 11 patents, and has buitl several award winning state of the art AI systems for businesses.

sk Kumar (Capital One)

Senthil Kumar is a Director of Data Science at Capital One where he applies Machine Learning and AI to various business problems. Prior to joining Capital One, he was at Bell Labs where he developed 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 KDD 2017 Workshop on Anomaly Detection in Finance, the 2018 NeurIPS Workshop on Challenges and Opportunities of AI in Financial Services: the Impact of Fairness, Explainability, Accuracy, and Privacy, the 2019 ICML Workshop on AI in Finance: Applications and Infrastructure for Multi-Agent Learning, and the 2019 2nd KDD Workshop on Anomaly Detection in Finance.

Isabelle Moulinier (Capital One)
Jiahao Chen (Capital One)

Jiahao Chen is a data science manager working in Capital One New York specializing in emerging technologies and university partnerships. He is currently the lead for the Banking in Explainable Algorithms Research (BEAR) group, focusing on FATES-related machine learning topics and their relation with banking regulations surrounding fair lending and explainability of credit decisioning to customers and regulators. Prior to joining Capital One in 2017, Jiahao was a Research Scientist at MIT CSAIL leading the Julia Lab, focusing on applications of the Julia programming language to various scientific data science problems and challenges in parallel computing and scientific computing.

John Paisley (Columbia University)

More from the Same Authors