Center for Digital Finance and Technology Announces New Research Grants
The Center for Digital Finance and Technology (CDFT) at Columbia Engineering is pleased to announce a second year of CDFT Research Grants beginning this Fall. Established in 2022, the Center is focused on advancing the digital transformation of financial services for higher efficiency and security, increased accessibility, and greater social responsibility.
As financial institutions have moved from local brick and mortar institutions, where the transaction took place in person, to multinational organizations serving a global community, the means of conducting financial business and operations have changed drastically, and demands for increased speed and scale of transactions and better security have become paramount. Added to these operational shifts is the recent advent of digital currencies and blockchain/smart-contract technologies.
Society is at a critical moment when understanding the potential of new financial technologies can make the difference between transformation and stagnation. The projects selected this year aim to understand the promises and pitfalls of these new financial innovations, and to improve their design to maximize adoption and accessibility.
Awarded proposals are:
Taming the Curse of Dimensionality in Blockchains: Design, Inference, and Economics.
PI: Wenpin Tang, Assistant Professor, Industrial Engineering and Operations Research
While decentralization is a unique feature of blockchains, it also results in high volumes of data. This project aims to improve blockchain design using big data, ultimately overcoming the “curse of dimensionality”. The research team will use a three-pronged approach, addressing the problem via algorithmic, statistical, and economic/social perspectives. They plan to devise novel scalable algorithms, perform dedicated analyses with tools including improved stochastic networks, anomaly detection, and high-dimensional robustness, and use the results to better understand economic incentives. This combined approach will help to create guidelines for policymakers and economists.
A Smart Contract-Based System for Lifetime Patch Management
PI: Simha Sethumadhavan, Professor, Computer Science
Attackers exploit software and hardware vulnerabilities, leading to severe consequences like data breaches and system failures. While patching these vulnerabilities is essential for maintaining security, it often results in system slowdowns and productivity losses, discouraging timely updates. This project proposes a compensation scheme based on the impact of patching, using smart contracts to automate the process. The system measures performance and frustration impacts, calculates compensation, and ensures prompt payouts. Additionally, it addresses privacy and regulatory concerns, providing a scalable, secure solution for ongoing patch management. The research team will prototype and validate these tools, ultimately aiming for widespread adoption and enhanced security.
Fair and Safe Marketplace Design for Online Video Games with Virtual Currencies
PI: Adam Elmachtoub, Associate Professor, Industrial Engineering and Operations Research
Many video games and virtual worlds include currencies that allow players to buy, sell, and trade goods, or even exchange for virtual for fiat currency. However, the resulting economic systems are little-understood by academia and by financial and regulatory agencies. This project investigates virtual video game marketplaces to better understand pricing, trade, and exchange mechanisms, with the goal of ensuring that these spaces are fair, free from arbitrage and predatory behavior, and maximize utility for platforms and players. The research team will develop a detailed model of an online world, which can be used to simulate gameplay and exchange, and eventually inform both research and policy on designing safe and fair games.
SMARTEST: A Multi-Modal LLM System for Effective Smart Contract Fuzzing
PI: Junfeng Yang, Professor, Computer Science
Smart contracts play a foundational role in the Web3 ecosystem, but are complex and prone to vulnerabilities that can be exploited by malicious actors. This project aims to apply fuzzing, a technique which can successfully identify vulnerabilities in traditional software, to smart contracts. The research team will build a multi-modal LLM system called SMARTEST, which will improve fuzzing’s performance on high-complexity and event-driven systems like smart contracts by identifying vulnerabilities deeply embedded in execution paths and understanding developers’ business and functional intents. The project will enhance the security of smart contracts, benefiting developers and the entire Web3 ecosystem.