用深度学习分析加密货币图表，发现简单图像更有效
Visual Chart Representations for Cryptocurrency Regime Prediction: A Systematic Deep Learning Study
- 对比三种图像编码方法与多种模型结构
- 4层CNN在原始图表上达0.892 AUC-ROC
- 小图+简单模型反而比大模型更优
技术交易者长期依赖蜡烛图的视觉分析来识别市场模式并预测价格走势。尽管深度学习在图像分类中取得显著成果，但其在金融图表图像上的应用仍较少被探索。本文系统比较了不同视觉表示方法在加密货币阶段预测中的表现。我们评估了三种图像编码方式（原始蜡烛图、格拉姆角场、多通道GAF）、五种图表组件配置、四种神经网络架构（CNN、ResNet18、EfficientNet-B0、Vision Transformer）以及ImageNet迁移学习的影响。通过在2018-2024年比特币、以太坊和标普500数据上进行八组受控实验，我们确定了视觉阶段分类的最佳配置。结果表明，基于原始蜡烛图的4层CNN达到0.892 AUC-ROC，优于更大规模的预训练模型。令人意外的是，更简单的表示（仅价格图表，128x128分辨率）始终优于更复杂的替代方案。我们利用GradCAM进行可解释性分析，并证明迁移学习虽存在领域差异，但仍能提升性能4%-16%。
原文摘要 · Abstract (English)
Technical traders have long relied on visual analysis of candlestick charts to identify market patterns and predict price movements. While deep learning has achieved remarkable success in image classification, its application to financial chart images remains underexplored. This paper presents a systematic study comparing different visual representations for cryptocurrency regime prediction. We evaluate three image encoding methods (raw candlestick charts, Gramian Angular Fields, and multi-channel GAF), five chart component configurations, four neural network architectures (CNN, ResNet18, EfficientNet-B0, and Vision Transformer), and the impact of ImageNet transfer learning. Through eight controlled experiments on Bitcoin, Ethereum, and S&P 500 data spanning 2018-2024, we identify optimal configurations for visual regime classification. Our results show that a simple 4-layer CNN on raw candlestick charts achieves 0.892 AUC-ROC, outperforming larger pretrained models. Surprisingly, simpler representations (price-only charts, 128x128 resolution) consistently outperform more complex alternatives. We provide interpretability analysis using GradCAM and demonstrate that transfer learning improves performance by 4-16% despite the domain gap between natural images and financial charts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。