Dreaming machine learning: Lipschitz extensions for reinforcement learning on financial markets

  title={Dreaming machine learning: Lipschitz extensions for reinforcement learning on financial markets},
  author={J. M. Calabuig and Herv{\'e} Falciani and Enrique Alfonso S{\'a}nchez-P{\'e}rez},

Figures from this paper

A Reinforcement Learning Based Encoder-Decoder Framework for Learning Stock Trading Rules
A novel end-to-end model based on the neural encoder-decoder framework combined with DRL is proposed to learn single instrument trading strategies from a long sequence of raw prices of the instrument.
Semi-Lipschitz functions and machine learning for discrete dynamical systems on graphs
The main objective is to explain the role of the lack of symmetry of quasi-metrics in the proposal: the irreversibility of dynamical processes is reflected in the asymmetry of their definition.
Analysing Stock Market Trend Prediction using Machine & Deep Learning Models: A Comprehensive Review
The applications of intelligent financial forecasting play an utmost important role in facilitating the investment decisions activities of many investors. With the right insight information, the
Self-defined information indices: application to the case of university rankings
A new and simple algorithm is presented to calculate an approximation of these indices using some standard bibliometric variables, such as the number of citations from the scientific output of universities and thenumber of articles per quartile.
Design Trend Forecasting by Combining Conceptual Analysis and Semantic Projections: New Tools for Open Innovation
A new trend analysis and forecasting method (Deflexor) is described, which is intended to help inform decisions in almost any field of human social activity, including, for example, business, art and design.
Evaluation Optimal Prediction Performance of MLMs on High-volatile Financial Market Data
The findings of study concludes that the algorithm of RF is most appropriate for nonlinear approximation/evaluation and the algorithms of SVR is most useful for high-frequency time-series data estimation.
Index spaces and standard indices in metric modelling
We analyze the basic structure of certain metric models, which are constituted by an index I acting on a metric space (D; d) representing a relevant property of the elements of D. We call such a
A Data Slicing Method to Improve Machine Learning Model Accuracy in Bankruptcy Prediction
  • Z. Ye
  • Computer Science
  • 2021
According to the findings in this research, the most related metric and the best variable to slice on to get a predictable sliced dataset turn out to be “Solvency Ratio” both in Chinese and Polish data.
Enhanced Food Safety Through Deep Learning for Food Recalls Prediction
A set of deep and machine learning techniques employing time series forecasting to provide insights regarding the risk associated with each product category concerning potential food recalls and an approach based on reinforcement learning which utilizes historical recall announcements for predicting future recalls that leads to timely recalls and contributes to enhanced food safety across the supply chain are introduced.


Deep Direct Reinforcement Learning for Financial Signal Representation and Trading
This work introduces a recurrent deep neural network for real-time financial signal representation and trading and proposes a task-aware backpropagation through time method to cope with the gradient vanishing issue in deep training.
Reinforcement Learning: An Introduction
This book provides a clear and simple account of the key ideas and algorithms of reinforcement learning, which ranges from the history of the field's intellectual foundations to the most recent developments and applications.
Graph kernels and Gaussian processes for relational reinforcement learning
This paper investigates the use of Gaussian processes to approximate the Q-values of state-action pairs in a relational setting and proposes graph kernels as a covariance function between state- action pairs.
Regime-switching recurrent reinforcement learning for investment decision making
It is argued that the RRL is unable to capture all the intricacies of financial time series, and proposed the RSRRL as a more suitable algorithm for such type of data.
Lipschitz Continuity in Model-based Reinforcement Learning
This work analyzes the impact of learning models that are Lipschitz continuous---the distance between function values for two inputs is bounded by a linear function of the distance between the inputs and proves an error bound for the value-function estimate arising from such models.
A Multiagent Approach to $Q$-Learning for Daily Stock Trading
A new stock trading framework is presented that incorporates multiple Q-learning agents, allowing them to effectively divide and conquer the stock trading problem by defining necessary roles for cooperatively carrying out stock pricing and selection decisions.
Algorithm Trading using Q-Learning and Recurrent Reinforcement Learning
This paper uses classic reinforcement algorithm, Q-learning, to evaluate the performance in terms of cumulative profits by maximizing different forms of value functions: interval profit, sharp ratio, and derivative sharp ratio and finds that this direct reinforcement learning framework enables a simpler problem representation than that in value function based search algorithm.