Adaptive Liquidity Provision in Uniswap V3 with Deep Reinforcement Learning

Haochen Zhang,Xi Chen,Lin F. Yang

2023-09-19

Abstract:Decentralized exchanges (DEXs) are a cornerstone of decentralized finance (DeFi), allowing users to trade cryptocurrencies without the need for third-party authorization. Investors are incentivized to deposit assets into liquidity pools, against which users can trade directly, while paying fees to liquidity providers (LPs). However, a number of unresolved issues related to capital efficiency and market risk hinder DeFi's further development. Uniswap V3, a leading and groundbreaking DEX project, addresses capital efficiency by enabling LPs to concentrate their liquidity within specific price ranges for deposited assets. Nevertheless, this approach exacerbates market risk, as LPs earn trading fees only when asset prices are within these predetermined brackets. To mitigate this issue, this paper introduces a deep reinforcement learning (DRL) solution designed to adaptively adjust these price ranges, maximizing profits and mitigating market risks. Our approach also neutralizes price-change risks by hedging the liquidity position through a rebalancing portfolio in a centralized futures exchange. The DRL policy aims to optimize trading fees earned by LPs against associated costs, such as gas fees and hedging expenses, which is referred to as loss-versus-rebalancing (LVR). Using simulations with a profit-and-loss (PnL) benchmark, our method demonstrates superior performance in ETH/USDC and ETH/USDT pools compared to existing baselines. We believe that this strategy not only offers investors a valuable asset management tool but also introduces a new incentive mechanism for DEX designers.

Artificial Intelligence

What problem does this paper attempt to address?

The paper attempts to address the issue of how to optimize liquidity provision strategies in decentralized exchanges (DEX), particularly in Uniswap V3, using deep reinforcement learning (DRL) to improve capital efficiency and reduce market risk.

Specifically, the paper focuses on:

1. **The contradiction between capital efficiency and market risk**: Uniswap V3 improves capital efficiency by allowing liquidity providers (LPs) to concentrate liquidity within specific price ranges. However, this approach also increases market risk because LPs can only earn trading fees when asset prices are within these preset ranges.

2. **Optimizing liquidity provision strategies**: To mitigate the above issue, the paper proposes a deep reinforcement learning-based method aimed at dynamically adjusting price ranges to maximize profits and reduce market risk. This method hedges liquidity positions by rebalancing portfolios on centralized futures exchanges, thereby neutralizing price movement risks.

3. **Loss and Rebalancing (LVR)**: The paper introduces a new metric—Loss and Rebalancing (LVR)—to evaluate the actual return on investment (ROI) of liquidity providers. LVR measures the value difference between the rebalanced portfolio and the original liquidity position, combined with trading fees and gas fees, providing a more realistic ROI assessment.

4. **Empirical analysis**: Through simulation experiments, the paper demonstrates that its method outperforms existing baseline methods in the ETH/USDC and ETH/USDT pools, especially in scenarios with limited initial capital.

In summary, the main goal of the paper is to provide a more effective strategy for liquidity providers in Uniswap V3 using deep reinforcement learning techniques to address the contradiction between market volatility and capital efficiency, thereby improving investment returns.