Research map / Reinforcement learning / Visual and representation RL
ArchetypeTrader: Reinforcement Learning for Selecting and Refining Learnable Strategic Archetypes in Quantitative Trading
Chuqiao Zong, Molei Qin, Haochong Xia, Bo An
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use VQ discrete archetype embeddings instead of continuous (ablation) | ArchetypeTrader (no refinement) · BTCUSDT | total return: -33.42 -> 57.47 | Improved |
| use VQ instead of continuous embeddings (ablation) | ArchetypeTrader (no refinement) · BNBUSDT | total return: 20.47 -> 21.56 | Improved |
| use VQ instead of k-means clustering (ablation) | ArchetypeTrader (no refinement) · DOTUSDT | total return: 36.84 -> 105.65 | Improved |
| add regret-aware refinement adapter (ablation) | ArchetypeTrader · DOTUSDT | total return: 105.65 -> 147.48 | Improved |
| add refinement adapter (ablation) | ArchetypeTrader · BTCUSDT | max drawdown: 19.91 -> 23.75 | Hurt |
| add refinement without regret penalty (ablation) | ArchetypeTrader · BNBUSDT | total return: 21.56 -> -4.29 | Hurt |
| add hindsight regret penalty to refinement (ablation) | ArchetypeTrader · BTCUSDT | total return: 30.24 -> 76.25 | Improved |
| horizon length 36 or 144 vs 72 (ablation) | ArchetypeTrader · DOTUSDT | total return: 147.48 -> 55.43 (h=36), 73.69 (h=144) | Hurt |
| codebook size 5 or 20 vs 10 (ablation) | ArchetypeTrader · DOTUSDT | total return: 147.48 -> 39.26 (K=5), 28.68 (K=20) | Hurt |
| ArchetypeTrader vs best baseline on BNBUSDT | ArchetypeTrader · BNBUSDT | total return: 88.32 (IV) vs 39.24 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- EarnHFT — EarnHFT: Efficient Hierarchical Reinforcement Learning for High Frequency Trading (AAAI 2024) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- EarnHFT: Efficient Hierarchical Reinforcement Learning for High Frequency Trading (AAAI 2024 · similarity 0.90)
- Logic-Q: Improving Deep Reinforcement Learning-based Quantitative Trading via Program Sketch-based Tuning (AAAI 2025 · similarity 0.88)
- Trade in Minutes! Rationality-Driven Agentic System for Quantitative Financial Trading (ICLR 2026 · similarity 0.86)
- AlphaPareto: Formulaic Alpha Discovery with LLM-Guided Multi-Objective Reinforcement Learning (NeurIPS 2026 · similarity 0.86)
- Robust Reinforcement Learning in Finance: Modeling Market Impact with Elliptic Uncertainty Sets (NeurIPS 2025 · similarity 0.85)
- AlphaQCM: Alpha Discovery in Finance with Distributional Reinforcement Learning (ICML 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
