Deep transformer q-learning based reinforcement learning for portfolio optimization of cryptocurrencies
1. 文件大小不能超过300M, 允许上传文档或压缩包等
2. 请确保上传文献的真实性、完整性，不得对原文做任何修改
注: 所有应助的资源仅供学习交流使用, 不得违反相关法律法规
DOI:
文献链接:
其他信息:
M Zepeda, W Kristjanpoller, MC Minutolo
Applied Soft Computing, 2026
Elsevier
Portfolio optimization remains a central theme in financial research, with recent advances in machine learning (ML) and computational power enabling more sophisticated forecasting models. Cryptocurrencies, characterized by extreme volatility and rapidly changing dynamics, pose particular challenges for portfolio allocation. This study proposes a novel Double Deep Transformer Q-Learning (DDTQL) framework for optimizing a portfolio of 15 leading cryptocurrencies. To the best of our knowledge, this is the first work to apply a …
