Summary
The financial industry leverages machine learning in more ways than just finding the right alpha signal. It grapples with supply chains, business processes, marketing, churn, fraud and money laundering, all while maintaining compliance with the various regulatory frameworks it is beholden to.
Due to the sheer volume of wealth being handled by the financial industry and its ubiquitous role in everyday life, it has been a lucrative target for a wide spectrum of ever-evolving bad actors. With each successive iteration of this workshop, we have attempted to capture the breadth of these actors - fraudsters, money launderers, market manipulators and potentially nation-state level risks.
With the advent of generative multimodal AI, the fusion of signals from conventional tabular datasets, time-series, free-text articles and earnings reports, images and networks, has de-siloed decision making to an unprecedented degree. This deluge of actionable information combined with easily available high performance commodity resources has significantly lowered the entry-barrier to using it in industry applications. We wish to explore the interplay of this breaking technology with an ever-evolving regulatory landscape.
GenAI offers groundbreaking approaches to handling the various data types prevalent in the financial sector. For tabular data, it can augment existing models by generating synthetic data for training, improving feature extraction, and enhancing predictive performance. In the realm of time series, Generative AI can model complex temporal patterns, offering more nuanced forecasts and risk assessments. For click streams and unstructured data, it can unearth patterns of customer behavior and preferences, enabling personalized services and fraud detection with unprecedented precision. While it is feasible to construct individual AI models tailored to each of these data types, achieving a comprehensive understanding of the customer necessitates the development of multimodal models capable of integrating all these data formats into an unified view. Generative AI holds the promise of providing effective solutions for this complex challenge
Decentralized finance (DeFi) has launched the fintech industry into Web 3.0, gained mainstream recognition and due skepticism. Along with this step-function change come several advantages of anonymity, speed and decoupling from fiat currency, but it is also laden with extreme risks and potential for criminal activity in a highly unregulated space. Through this workshop, we would like to continue to explore the applications of machine learning in this rapidly evolving domain decoupled from the hype-cycle.
The purpose of this workshop is to bring together researchers and practitioners to discuss both the problems faced by the financial industry and potential solutions. We invite regular papers, positional papers and extended abstracts of work in progress. We also encourage short papers from financial industry practitioners that introduce domain specific problems and challenges to academic researchers. This event will be the seventh in a sequence of finance related workshops we have organized at KDD. The first workshop was held at KDD 2017, the second workshop at KDD 2019, the third workshop at KDD 2020 , the fourth workshop at KDD 2021, the fifth workshop at KDD 2022 and the sixth at KDD 2023.
Call for Papers
We invite papers on machine learning and AI with applications to the financial industry. Topics of interest include, but are not limited to, the following:
Mining for signals in financial data
Vetting and sourcing data for high-stakes decision making
Transfer learning from Time Series, Recommendation Systems, Reinforcement Learning, Network Science, Image Processing etc
Patterns and anti-patterns in early-detection
Multi modal machine learning in practice
Sensor fusion approaches in the use of alternative data
Use cases like marketing, anomaly detection, churn prevention, etc
Generating synthetic data for privacy preservation, data sharing, robustness, etc
Generative AI
Foundational models for financial data
Novel applications of generative techniques in the financial sector
Fine-tuning, FSL, ZSL on off-the-shelf foundational models - fairness, bias and explainability implications
Threat response to AI-generated attacks
Regulatory implications to the use of generative AI
Detecting bad actors and activity in financial data
Detecting anomalies at large in unsupervised/semi-supervised settings
Active learning strategies in noisy and uncertain environments. Reinforcement learning strategies and their applications to gather ground truth
Model calibration, stability and adaptiveness trade-offs
Insider trading
Fraud and abuse
Cyber threats
Money Laundering
Compliance violations
Model Explainability & Governance
Role of explainability in several verticals, markets
Deployed and vetted applications with explainability
Fairness
Fairness in the context of finance - lending and beyond!
Privacy preservation
Reassessing credit in the conventional sense
Role of DeFi in fairness
Market Manipulation
Analysis of limit order book feeds
Fake news and other noisy social signal ingestion challenges
Robustness to adversarial actors
Crypto and DeFi
Specific challenges and analysis of high risk domains
Best practices to thwart bad actors and stay compliant with an in-flux regulatory landscape
We also invite tutorials and introductory papers to bridge the gap between academia and the financial industry:
Overview of Industry Challenges
Short papers from financial industry practitioners that introduce domain specific problems and challenges to academic researchers. These papers should describe problems that can inspire new research directions in academia, and should serve to bridge the information gap between academia and the financial industry.
Algorithmic Tutorials
Short tutorials from academic researchers that explain current solutions to challenges related to the technical areas mentioned above, not necessarily limited to the financial domain. These tutorials will serve as an introduction and enable financial industry practitioners to employ/adapt latest academic research to their use-cases.
Submission Guidelines
All submissions must be PDFs formatted in the Standard ACM Conference Proceedings Template. Submissions are limited to 8 content pages or less, including all figures, tables and appendices but excluding references. All accepted papers will be presented as posters and some would be selected for oral presentations, depending on schedule constraints. Accepted papers will be posted on the workshop website.
Following the KDD conference submission policy, reviews are double-blind, and author names and affiliations should NOT be listed.
Papers should be submitted on the submission portal by May 28, 2024 11:59 PM Pacific Time
Key Dates
Submission Deadline: May 28, 2024, 11:59 Pacific Time
Author Notification: June 28, 2024
Camera ready papers due: July 30, 2024
Workshop: August 25-29, 2024