Trading de criptomonedas mediante algoritmos de aprendizaje por refuerzo profundo y aprendizaje automático
Palabras clave:
aprendizaje automático, criptomonedas, trading, aprendizaje por refuerzo.Resumen
El objetivo de este artículo es realizar una comparación del desempeño de algoritmos de aprendizaje automático y de aprendizaje por refuerzo profundo para la compra y venta de criptomonedas basándose en la predicción de precios. Esto se lleva a cabo mediante el entrenamiento y prueba de dichos modelos usando información diaria que abarca el periodo del 2017-11-09 al 2022-05-01 y fue descargada de la página de Yahoo finance en su sección de criptomonedas. Se observa que los algoritmos de aprendizaje automático tienen métricas de precisión altas en sus predicciones y se ejecutan de forma rápida, aunque no resultan en alto retorno de la inversión, ni tampoco sobrepasan al aprendizaje por refuerzo profundo en este sentido.
Referencias
Alessandretti, L., ElBahrawy, A., Aiello, L. M., & Baronchelli, A. (2018). Anticipating cryptocurrency prices using machine learning. Complexity, 2018, 1-16. Accedido 07-03-2021.
Bancomer, B. (2021). 'trading'? Obtenido de https://www.bbva.com/es/que-es-trading-que-hace-falta-para-operar/. Accedido 21-02-2021.
Brownlee, J. (2017). Autoregression models for time series forecasting with python. Machine Learning Mastery, 2(01). Obtenido de https://machinelearningmastery.com/autoregressionmodels-time-series-forecasting-python/. Accedido 06-07-2022.
Bu, S. J., & Cho, S. B. (2018). Learning optimal Q-function using deep Boltzmann machine for reliable trading of cryptocurrency. In Intelligent Data Engineering and Automated Learning–IDEAL 2018: 19th International Conference, Madrid, Spain, November 21–23, 2018, Proceedings, Part I 19 (pp. 468-480). Springer International Publishing. Accedido 22-02-2021.
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794). Accedido 01-12-2022.
Choudhary, A. (2019). A hands-on introduction to deep q-learning using openai gym in python. Analytics Vidhya. Obtenido de https://www.analyticsvidhya.com/blog/2019 / 04/introduction-deep-qlearning-python/. Accedido 27-05-2021.
DeepRobotics (2016). A comprehensive approach to reinforcement learning. Obtenido de https://vmayoral.github.io/robots,/ai,/deep/learning,/rl,/reinforcement/learning tutorial/. Accedido 28-05-2021.
Fradkov, A. L. (2020). Early history of machine learning. IFAC-PapersOnLine, 53(2), 1385-1390.
GreatLearning (2020). Understanding xgboost algorithm I what is xgboost algorithm? Obtenido de https://www.mygreatlearning.com/blog/xgboostalgorithm/. Accedido 05-07-2022.
Gu, S., Lillicrap, T., Sutskever, I., & Levine, S. (2016). Continuous deep q-learning with model-based acceleration. In International conference on machine learning (pp. 2829-2838). PMLR. Accedido 24-02-2021.
Guo, J. and Tuckfield, B. (2020). News-based machine learning and deep learning methods for stock prediction. Journal of Physics: Conference Series, 1642:012014. Accedido 15-03-2022.
Gupta, S. (2018). Sentiment analysis: Concept, analysis and applications. Toward Data Science. Obtenido de https://towardsdatascience.com/sentiment-analysis-concept analysis-and-applications-6c94d6f58c17. Accedido 25-02-2021.
IBM (2020). Machine learning. Obtenido de https://www.ibm.com/cloud/learn/machine-learning. Accedido 12-04-2022.
Jiang, Z., & Liang, J. (2017). Cryptocurrency portfolio management with deep reinforcement learning. In 2017 Intelligent systems conference (IntelliSys) (pp. 905-913). IEEE. Accedido 22-02-2021.
Juchli, M. (2018). Limit order placement optimization with deep reinforcement learning: Learning from patterns in cryptocurrency market data. Accedido 2402-2021.
Julián, G. (2016). Las redes neuronales: qué son y por qué están volviendo. Revista Xataka. Obtenido de https://www.xataka.com/robotica-e-ia/las-redes-neuronales-queson-y-por-que-estan-volviendo. Accedido 20-05-2021.
Keller, A., & Scholz, M. (2019). Trading on cryptocurrency markets: Analyzing the behavior of bitcoin investors. Accedido 28-02-2021.
Kolla, B. (2020). Predicting crypto currency prices using machine learning and deep learning techniques. International Journal of Advanced Trends in Computer Science and Engineering, 9. Accedido 22-02-2021.
Liang, Z., Chen, H., Zhu, J., Jiang, K., & Li, Y. (2018). Adversarial deep reinforcement learning in portfolio management. arXiv preprint arXiv:1808.09940. Accedido 2202-2021.
Lucarelli, G., & Borrotti, M. (2019). A deep reinforcement learning approach for automated cryptocurrency trading. In Artificial Intelligence Applications and Innovations: 15th IFIP WG 12.5 International Conference, AIAI 2019, Hersonissos, Crete, Greece, May 24–26, 2019, Proceedings 15 (pp. 247-258). Springer International Publishing. Accedido 21-02-2021.
Lucarelli, G., & Borrotti, M. (2020). A deep Q-learning portfolio management framework for the cryptocurrency market. Neural Computing and Applications, 32, 17229-17244. Accedido 21-02-2021.
Lutins, E. (2017). Grid searching in machine learning: Quick explanation and python implementation. Medium, Medium, 5. Obtenido de https://elutins.medium.com/gridsearching-in-machine-learning-quick-explanation-and-pythonimplementation-550552200596. Accedido 07-07-2022.
McNally, S., Roche, J., & Caton, S. (2018, March). Predicting the price of bitcoin using machine learning. In 2018 26th euromicro international conference on parallel, distributed and network-based processing (PDP) (pp. 339-343). IEEE. Accedido 22-02-2021.
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A., Veness, J., Bellemare, M., Graves, A., Riedmiller, M., Fidjeland, A., Ostrovski, G., Petersen, S., Beattie, C., Sadik, A., Antonoglou, I., King, H., Kumaran, D., Wierstra, D., Legg, S., and Hassabis, D. (2015). Human-level control through deep reinforcement learning. Nature, 518:529-33. Accedido 27-05-2021.
Morales, M. (2019). Deep Reinforcement Learning. Accedido 2-12-2021.
Patel, Y. (2018). Optimizing market making using multi-agent reinforcement learning. arXiv preprint arXiv:1812.10252. Accedido 22-02-2021.
Restuputri, D.P.; Refoera, F.B.; Masudin, I. Investigating Acceptance of Digital Asset and Crypto Investment Applications Based on the Use of Technology Model (UTAUT2). FinTech 2023, 2, 388-413. https://doi.org/10.3390/fintech2030022
Sabry, F., Labda, W., Erbad, A., and Malluhi, Q. (2020). Cryptocurrencies and artificial intelligence: Challenges and opportunities. IEEE Access, 8: 175840 175858. Accedido 21-02-2021.
Salakhutdinov, R. and Hinton, G. (2009). Deep Boltzmann Machines, volume 5 of Proceedings of Machine Learning Research. Hilton Clearwater Beach Resort, Clearwater Beach, Florida USA. Accedido 24-02-2021.
Sen, J. (2022). Machine learning and deep learning in stock price prediction. Machine Learning in the Analysis and Forecasting of Financial Time Series, 29-67. Accedido 15-03-2022.
Shubham, J. (2017). A comprehensive beginners guide for linear, ridge and lasso regression in python and r. Obtenido de https://www.analyticsvidhya.com/blog/2017/06/a-comprehensive-guidefor-linear-ridge-and-lasso-regression/h2_5. Accedido 14-04-2022.
Shukla, L. (2019). Fundamentals of neural networks. Obtenido de https://wandb.ai/site/articles/fundamentals-of-neural-networks. Accedido 2105-2021.
Soni, S. (2021). Crypto rally: Total market cap hits new all-time high of 2,8t;1t added in just over a months time. Obtenido de https://www.financialexpress.com/market/crypto-rally-total-market-caphits-new-all-time-high-of-2-8t-1 t-added-in-just-over-a-months-time /2362504/. Accedido 25-11-2021.
Sruthi, E. (2021). Understanding random forest. Obtenido de https://www.analyticsvidhya.com/blog/2021/06/understanding-randomforest/. Accedido 05-07-2022.
Torres, J. (2020). A gentle introduction to deep reinforcement learning. Obtenido de https://towardsdatascience.com/drl-01-a-gentle-introduction-todeep-reinforcement-learning-405b79866bf4. Accedido 20-05-2021.
Torres, J. (2021). Introducción al aprendizaje por refuerzo profundo. Accedido 10-112021.
Trozze et al. (2022). Cryptocurrencies and future financial crime. Crime Science. https://doi.org/10.1186/s40163-021-00163-8.
Vadapalli, P. (2020). Bagging vs boosting in machine learning: Difference between bagging and boosting. Obtenido de https://www.upgrad.com/blog/bagging vs-boosting. Accedido 05-07-2022.
Youngson, M. (2020). Introduction to reinforcement learning. Obtenido de https://medium.com/swlh/introduction-to-reinforcement-learning63ff8923bd88. Accedido 27-05-2021.
Zhang, A. (2021). How to design a reinforcement learning reward function for a lunar lander. Obtenido de https://towardsdatascience.com/how-to-designreinforcement-learning-reward-function-for-a-lunar-lander-562a24c393f6. Accedido 2-12-2021.
