Research map / Image generation / Fast diffusion sampling
Controllable Financial Market Generation with Diffusion Guided Meta Agent
Yu-Hao Huang, Chang Xu, Yang Liu, Weiqing Liu, Wu-Jun Li, Jiang Bian
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 |
|---|---|---|---|
| add continuous control conditioning (ablation) | DigMA · A-Main Return Lower bin | MSE: 1.443 -> 0.206 | Improved |
| add continuous control conditioning (ablation) | DigMA · A-Main Volatility Higher bin | MSE: 4.288 -> 2.389 | Improved |
| use continuous instead of discrete control encoder | DigMA · A-Main Return Lower bin | MSE: 1.055 -> 0.206 | Improved |
| use continuous instead of discrete control encoder | DigMA · ChiNext Volatility Higher bin | MSE: 2.257 -> 2.631 | Hurt |
| use DigMA instead of LOBGAN | LOBGAN · A-Main | K-L VolC: 1.101 -> 0.273 | Improved |
| use DigMA instead of LOBGAN on return autocorrelation | LOBGAN · A-Main | K-L RetAC: 1.903 -> 2.781 | Hurt |
| train RL agent in DigMA environment instead of replay | RL trader · HFT RL | Sharpe ratio: 0.014 -> 0.049 | Improved |
| enable meta-controller conditioning (DigMA vs DigMA-c) (ablation) | RL trader · HFT RL | Max drawdown: 0.715 -> 1.313 | Hurt |
| enable meta-controller conditioning (DigMA vs DigMA-c) (ablation) | RL trader · HFT RL | Daily return: 0.015 -> 0.029 | Improved |
| use DigMA instead of LOBGAN | LOBGAN · Generation latency | Time (ms) per order: 1.710 -> 0.017 | Improved |
Closest papers by meaning
- MarS: a Financial Market Simulation Engine Powered by Generative Foundation Model (ICLR 2025 · similarity 0.89)
- Trade in Minutes! Rationality-Driven Agentic System for Quantitative Financial Trading (ICLR 2026 · similarity 0.85)
- R&D-Agent-Quant: A Multi-Agent Framework for Data-Centric Factors and Model Joint Optimization (NeurIPS 2025 · similarity 0.85)
- TwinMarket: A Scalable Behavioral and Social Simulation for Financial Markets (NeurIPS 2025 · similarity 0.84)
- Market-GAN: Adding Control to Financial Market Data Generation with Semantic Context (AAAI 2024 · similarity 0.83)
- EarnHFT: Efficient Hierarchical Reinforcement Learning for High Frequency Trading (AAAI 2024 · similarity 0.83)
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.
