Summary

One of the fundamental promises of deep learning is its ability to build up increasingly meaningful representations of the data from complex but raw inputs. These techniques demonstrate remarkable efficacy on high dimensional data with unique proximity structures (image, natural language, graphs). These representations have proven to be useful not just for the task they are trained on but as general feature learners for a variety of downstream tasks.

Not only are these types of data prevalent in financial services and e-commerce, but also they often capture extremely interesting aspects of social and economic behavior. For example, both financial transactions and online purchases can be viewed as edges on graphs of economic activity. To date, these graphs are far less studied than social networks, though they provide a unique look at behavior, social structures, and risk. Meanwhile, activity or transaction sequences, usually determined by user sessions, can reflect the users’ long term and short term interests, which can be modeled by sequential models, and used to predict the user’s future activities. Although language models have been explored in session data modeling, how to re-use the representations learned from one job to another job effectively is still an open question.

With the abundance of these rich datasets amenable to representation learning, the space for applications is quite large. This can include graph representation learning for fraud detection, session embeddings for customer servicing and messaging, item and query representation for search and retrieval, item and customer representation for personalization, etc.

The aim of this workshop is to bring together researchers from these different domains to discuss the application of representation learning to financial services and e-commerce applications. For the first time, researchers from four major e-commerce companies (Amazon, Walmart, Alibaba and eBay) and two banks (JP Morgan and Capital One) have come together to organize this workshop along with researchers from academia. A shared goal across these industries and application areas is to transform large-scale representational data into tangible revenue for businesses. Towards this goal, our confirmed invited speakers will share diverse perspectives on ways that representation learning can be used to solve problems in financial services and e-commerce. This will also be a forum to share how research on financial services and e-commerce data provides unique insights into socio-economic behavior. Lastly, it will be an opportunity to share the novel challenges presented by the types of data collected in financial services.

Call for Papers

We invite short papers on all aspects of representation learning related to financial services and e-commerce.

Topics include (but not limited to) the following:

• Representation learning on financial graphs

• Representation learning in language models

• Representation learning on tabular datasets

• Event sequence representations

• Multi-modal fusion for combining disparate data types

• Dynamics of representations over time

• Representation learning for anomaly detection

• Gleaning behavioral, social, and economic insights from representations of financial datasets

• Identifying and mitigating bias in representations

Applications covered in this workshop will include but not limited to:

• Finance

• E-commerce

• Fraud detection

• Financial planning

• Econometric models

• Economic forecasting

• Recommender Systems

• Asset valuation

• Financial instrument pricing

• Algorithmic trading

• Loan underwriting

• Portfolio management

• Risk management

• Customer service

• Privacy

We also invite tutorials and introductory papers to bridge the gap between academia and the financial industry, and position papers that provide interdisciplinary perspectives:

• 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 representation learning 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 ICML style. Submissions are limited to 8 content pages, including all figures and tables but excluding references. Despite this page limit, we also welcome and encourage short papers (2-4 pages) to be submitted. All accepted papers will be presented as posters; some may be selected for highlights or contributed talks, depending on schedule constraints. Accepted papers will be posted on the workshop website.

Papers should be submitted on CMT3 by May 31, 2021 23:59 AoE

https://cmt3.research.microsoft.com/RLFECA2021

Key dates

• Submission deadline: May 31, 2021 23:59 AoE at https://cmt3.research.microsoft.com/RLFECA2021

• Author notification: June 18, 2021

• Authors of accepted papers must submit video recordings of their talks by June 27, 2021

• Camera-ready papers due: July 15, 2021

• Workshop: July 23, 2021