{"id":"arxiv:1807.02787","title":"Financial Trading as a Game: A Deep Reinforcement Learning Approach","date":"2018-07-08","section":"Finance","quality":-0.745254,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/","source_url":"https://arxiv.org/abs/1807.02787","text_available":true,"authors":["Chien Yi Huang"],"outgoing":[{"source":"arxiv:1807.02787","target":"arxiv:1707.06887","evidence":"id","id":"arxiv:1707.06887","title":"A Distributional Perspective on Reinforcement Learning","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/96b040bb771d28ca3254b5519a2191c3e958a35dfdd60075344ac86af6a590e1/"}],"incoming":[],"text_parts":[{"start":0,"end":15918,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html"},{"start":15918,"end":26317,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html"}],"contents":[{"title":"Financial Trading as a Game: A Deep Reinforcement Learning Approach","offset":0,"level":1,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"Abstract","offset":71,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"keywords","offset":1627,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"1 Introduction","offset":1644,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"1.1 The Financial Trading Task","offset":1972,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"1.2 Motivation","offset":3904,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"Definition 1 .","offset":4456,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"1.3 Challenges","offset":5069,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"1.4 Contributions","offset":6652,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2 Method","offset":8553,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.1 Data Preparation and Feature Extraction","offset":8696,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.2 Financial Trading MDP","offset":9258,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.2.1 State Space ∈ ℝ 198 \\in\\mathbb{R}^{198}","offset":9409,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.2.2 Action Space","offset":10828,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.2.3 Reward Function","offset":11330,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.3 Fully Exploit with Action Augmentation","offset":12778,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.4 Model Architecture","offset":14688,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.4.1 Weight Initialization","offset":15119,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.5 Training Scheme","offset":15697,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.5.1 Modified Training Scheme","offset":15881,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-1.html#text"},{"title":"2.5.2 A Complete Online Learning Algorithm","offset":17287,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"3 Experiment","offset":18156,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"3.1 Hyperparameters","offset":18405,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"3.2 Simulation Result","offset":19125,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"3.3 Effect of the Spread","offset":21106,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"3.4 Effectiveness of Action Augmentation","offset":22123,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"4 Conclusion","offset":22622,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"4.1 Achievements","offset":22713,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"4.2 Future Work","offset":24056,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"},{"title":"Appendix","offset":26223,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/part-2.html#text"}],"text_code_points":26317,"source_markdown":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ffab71fb6c4e10245324b584ff3252a6fae8005a8220f84f1c44da083caec855/source.md"}