Reinforcement Learning

Minimal Shortfall Strategies for Liquidation of a Basket of Stocks using Reinforcement Learning ArXiv ID: 2502.07868 “View on arXiv” Authors: Unknown Abstract This paper studies the ubiquitous problem of liquidating large quantities of highly correlated stocks, a task frequently encountered by institutional investors and proprietary trading firms. Traditional methods in this setting suffer from the curse of dimensionality, making them impractical for high-dimensional problems. In this work, we propose a novel method based on stochastic optimal control to optimally tackle this complex multidimensional problem. The proposed method minimizes the overall execution shortfall of highly correlated stocks using a reinforcement learning approach. We rigorously establish the convergence of our optimal trading strategy and present an implementation of our algorithm using intra-day market data. ...

Regret-Optimized Portfolio Enhancement through Deep Reinforcement Learning and Future Looking Rewards ArXiv ID: 2502.02619 “View on arXiv” Authors: Unknown Abstract This paper introduces a novel agent-based approach for enhancing existing portfolio strategies using Proximal Policy Optimization (PPO). Rather than focusing solely on traditional portfolio construction, our approach aims to improve an already high-performing strategy through dynamic rebalancing driven by PPO and Oracle agents. Our target is to enhance the traditional 60/40 benchmark (60% stocks, 40% bonds) by employing the Regret-based Sharpe reward function. To address the impact of transaction fee frictions and prevent signal loss, we develop a transaction cost scheduler. We introduce a future-looking reward function and employ synthetic data training through a circular block bootstrap method to facilitate the learning of generalizable allocation strategies. We focus on two key evaluation measures: return and maximum drawdown. Given the high stochasticity of financial markets, we train 20 independent agents each period and evaluate their average performance against the benchmark. Our method not only enhances the performance of the existing portfolio strategy through strategic rebalancing but also demonstrates strong results compared to other baselines. ...

Minimal Batch Adaptive Learning Policy Engine for Real-Time Mid-Price Forecasting in High-Frequency Trading ArXiv ID: 2412.19372 “View on arXiv” Authors: Unknown Abstract High-frequency trading (HFT) has transformed modern financial markets, making reliable short-term price forecasting models essential. In this study, we present a novel approach to mid-price forecasting using Level 1 limit order book (LOB) data from NASDAQ, focusing on 100 U.S. stocks from the S&P 500 index during the period from September to November 2022. Expanding on our previous work with Radial Basis Function Neural Networks (RBFNN), which leveraged automated feature importance techniques based on mean decrease impurity (MDI) and gradient descent (GD), we introduce the Adaptive Learning Policy Engine (ALPE) - a reinforcement learning (RL)-based agent designed for batch-free, immediate mid-price forecasting. ALPE incorporates adaptive epsilon decay to dynamically balance exploration and exploitation, outperforming a diverse range of highly effective machine learning (ML) and deep learning (DL) models in forecasting performance. ...

INVESTORBENCH: A Benchmark for Financial Decision-Making Tasks with LLM-based Agent ArXiv ID: 2412.18174 “View on arXiv” Authors: Unknown Abstract Recent advancements have underscored the potential of large language model (LLM)-based agents in financial decision-making. Despite this progress, the field currently encounters two main challenges: (1) the lack of a comprehensive LLM agent framework adaptable to a variety of financial tasks, and (2) the absence of standardized benchmarks and consistent datasets for assessing agent performance. To tackle these issues, we introduce \textsc{“InvestorBench”}, the first benchmark specifically designed for evaluating LLM-based agents in diverse financial decision-making contexts. InvestorBench enhances the versatility of LLM-enabled agents by providing a comprehensive suite of tasks applicable to different financial products, including single equities like stocks, cryptocurrencies and exchange-traded funds (ETFs). Additionally, we assess the reasoning and decision-making capabilities of our agent framework using thirteen different LLMs as backbone models, across various market environments and tasks. Furthermore, we have curated a diverse collection of open-source, multi-modal datasets and developed a comprehensive suite of environments for financial decision-making. This establishes a highly accessible platform for evaluating financial agents’ performance across various scenarios. ...

Multimodal Deep Reinforcement Learning for Portfolio Optimization ArXiv ID: 2412.17293 “View on arXiv” Authors: Unknown Abstract We propose a reinforcement learning (RL) framework that leverages multimodal data including historical stock prices, sentiment analysis, and topic embeddings from news articles, to optimize trading strategies for SP100 stocks. Building upon recent advancements in financial reinforcement learning, we aim to enhance the state space representation by integrating financial sentiment data from SEC filings and news headlines and refining the reward function to better align with portfolio performance metrics. Our methodology includes deep reinforcement learning with state tensors comprising price data, sentiment scores, and news embeddings, processed through advanced feature extraction models like CNNs and RNNs. By benchmarking against traditional portfolio optimization techniques and advanced strategies, we demonstrate the efficacy of our approach in delivering superior portfolio performance. Empirical results showcase the potential of our agent to outperform standard benchmarks, especially when utilizing combined data sources under profit-based reward functions. ...

Financial News-Driven LLM Reinforcement Learning for Portfolio Management ArXiv ID: 2411.11059 “View on arXiv” Authors: Unknown Abstract Reinforcement learning (RL) has emerged as a transformative approach for financial trading, enabling dynamic strategy optimization in complex markets. This study explores the integration of sentiment analysis, derived from large language models (LLMs), into RL frameworks to enhance trading performance. Experiments were conducted on single-stock trading with Apple Inc. (AAPL) and portfolio trading with the ING Corporate Leaders Trust Series B (LEXCX). The sentiment-enhanced RL models demonstrated superior net worth and cumulative profit compared to RL models without sentiment and, in the portfolio experiment, outperformed the actual LEXCX portfolio’s buy-and-hold strategy. These results highlight the potential of incorporating qualitative market signals to improve decision-making, bridging the gap between quantitative and qualitative approaches in financial trading. ...

The Role of AI in Financial Forecasting: ChatGPT’s Potential and Challenges ArXiv ID: 2411.13562 “View on arXiv” Authors: Unknown Abstract The outlook for the future of artificial intelligence (AI) in the financial sector, especially in financial forecasting, the challenges and implications. The dynamics of AI technology, including deep learning, reinforcement learning, and integration with blockchAIn and the Internet of Things, also highlight the continued improvement in data processing capabilities. Explore how AI is reshaping financial services with precisely tAIlored services that can more precisely meet the diverse needs of individual investors. The integration of AI challenges regulatory and ethical issues in the financial sector, as well as the implications for data privacy protection. Analyze the limitations of current AI technology in financial forecasting and its potential impact on the future financial industry landscape, including changes in the job market, the emergence of new financial institutions, and user interface innovations. Emphasizing the importance of increasing investor understanding and awareness of AI and looking ahead to future trends in AI tools for user experience to drive wider adoption of AI in financial decision making. The huge potential, challenges, and future directions of AI in the financial sector highlight the critical role of AI technology in driving transformation and innovation in the financial sector ...

Reinforcement Learning Methods for the Stochastic Optimal Control of an Industrial Power-to-Heat System ArXiv ID: 2411.02211 “View on arXiv” Authors: Unknown Abstract The optimal control of sustainable energy supply systems, including renewable energies and energy storage, takes a central role in the decarbonization of industrial systems. However, the use of fluctuating renewable energies leads to fluctuations in energy generation and requires a suitable control strategy for the complex systems in order to ensure energy supply. In this paper, we consider an electrified power-to-heat system which is designed to supply heat in form of superheated steam for industrial processes. The system consists of a high-temperature heat pump for heat supply, a wind turbine for power generation, a sensible thermal energy storage for storing excess heat and a steam generator for providing steam. If the system’s energy demand cannot be covered by electricity from the wind turbine, additional electricity must be purchased from the power grid. For this system, we investigate the cost-optimal operation aiming to minimize the electricity cost from the grid by a suitable system control depending on the available wind power and the amount of stored thermal energy. This is a decision making problem under uncertainties about the future prices for electricity from the grid and the future generation of wind power. The resulting stochastic optimal control problem is treated as finite-horizon Markov decision process for a multi-dimensional controlled state process. We first consider the classical backward recursion technique for solving the associated dynamic programming equation for the value function and compute the optimal decision rule. Since that approach suffers from the curse of dimensionality we also apply reinforcement learning techniques, namely Q-learning, that are able to provide a good approximate solution to the optimization problem within reasonable time. ...

FinVision: A Multi-Agent Framework for Stock Market Prediction ArXiv ID: 2411.08899 “View on arXiv” Authors: Unknown Abstract Financial trading has been a challenging task, as it requires the integration of vast amounts of data from various modalities. Traditional deep learning and reinforcement learning methods require large training data and often involve encoding various data types into numerical formats for model input, which limits the explainability of model behavior. Recently, LLM-based agents have demonstrated remarkable advancements in handling multi-modal data, enabling them to execute complex, multi-step decision-making tasks while providing insights into their thought processes. This research introduces a multi-modal multi-agent system designed specifically for financial trading tasks. Our framework employs a team of specialized LLM-based agents, each adept at processing and interpreting various forms of financial data, such as textual news reports, candlestick charts, and trading signal charts. A key feature of our approach is the integration of a reflection module, which conducts analyses of historical trading signals and their outcomes. This reflective process is instrumental in enhancing the decision-making capabilities of the system for future trading scenarios. Furthermore, the ablation studies indicate that the visual reflection module plays a crucial role in enhancing the decision-making capabilities of our framework. ...

Reinforcement Learning in Non-Markov Market-Making ArXiv ID: 2410.14504 “View on arXiv” Authors: Unknown Abstract We develop a deep reinforcement learning (RL) framework for an optimal market-making (MM) trading problem, specifically focusing on price processes with semi-Markov and Hawkes Jump-Diffusion dynamics. We begin by discussing the basics of RL and the deep RL framework used, where we deployed the state-of-the-art Soft Actor-Critic (SAC) algorithm for the deep learning part. The SAC algorithm is an off-policy entropy maximization algorithm more suitable for tackling complex, high-dimensional problems with continuous state and action spaces like in optimal market-making (MM). We introduce the optimal MM problem considered, where we detail all the deterministic and stochastic processes that go into setting up an environment for simulating this strategy. Here we also give an in-depth overview of the jump-diffusion pricing dynamics used, our method for dealing with adverse selection within the limit order book, and we highlight the working parts of our optimization problem. Next, we discuss training and testing results, where we give visuals of how important deterministic and stochastic processes such as the bid/ask, trade executions, inventory, and the reward function evolved. We include a discussion on the limitations of these results, which are important points to note for most diffusion models in this setting. ...