Prior Knowledge Neural Network for Automatic Feature Construction in Financial Time Series

In quantitative finance, useful features are constructed by human experts. However, this method is of low efficient. Thus, automatic feature construction algorithms have received more and more attention. The SOTA technic in this field is to use reverse polish expression to represent the features, and then use genetic programming to reconstruct it. In this paper, we propose a new method, alpha discovery neural network, which can automatically construct features by using neural network. In this work, we made several contributions. Firstly, we put forward new object function by using empirical knowledge in financial signal processing, and we also fixed its undifferentiated problem. Secondly, we use model stealing technic to learn from other prior knowledge, which can bring enough diversity into our network. Thirdly, we come up with a method to measure the diversity of different financial features. Experiment shows that ADN can produce more diversified and higher informative features than GP. Besides, if we use GP's output to serve as prior knowledge, its final achievements will be significantly improved by using ADN.

[1]  Jimeng Sun,et al.  Opportunities and challenges in developing deep learning models using electronic health records data: a systematic review , 2018, J. Am. Medical Informatics Assoc..

[2]  Ferat Sahin,et al.  A survey on feature selection methods , 2014, Comput. Electr. Eng..

[3]  Kenneth DeJong,et al.  Genetic algorithms as a tool for restructuring feature space representations , 1995, Proceedings of 7th IEEE International Conference on Tools with Artificial Intelligence.

[4]  Marianthi Markatou,et al.  Text mining for the Vaccine Adverse Event Reporting System: medical text classification using informative feature selection , 2011, J. Am. Medical Informatics Assoc..

[5]  Parikshit Sondhi,et al.  Feature Construction Methods : A Survey , 2009 .

[6]  Zura Kakushadze,et al.  101 Formulaic Alphas , 2015, 1601.00991.

[7]  Bing Xue,et al.  Genetic Programming for Region Detection, Feature Extraction, Feature Construction and Classification in Image Data , 2016, EuroGP.

[8]  Peter A. N. Bosman,et al.  Symbolic regression and feature construction with GP-GOMEA applied to radiotherapy dose reconstruction of childhood cancer survivors , 2018, GECCO.

[9]  Mengjie Zhang,et al.  Genetic programming for feature construction and selection in classification on high-dimensional data , 2016, Memetic Comput..

[10]  James M. Schombert,et al.  The LkH-alpha 101 infrared cluster , 1991 .

[11]  Mengjie Zhang,et al.  Genetic programming for improving image descriptors generated using the scale-invariant feature transform , 2012, IVCNZ '12.

[12]  Anne E Carpenter,et al.  Opportunities and obstacles for deep learning in biology and medicine , 2017, bioRxiv.

[13]  Zhaojun Yang,et al.  Machine Learning Solutions to Challenges in Finance: An Application to the Pricing of Financial Products , 2020 .

[14]  Jason H. Moore,et al.  A General Feature Engineering Wrapper for Machine Learning Using \epsilon -Lexicase Survival , 2017, EuroGP.

[15]  Mengjie Zhang,et al.  Wrapper Feature Construction for Figure-Ground Image Segmentation Using Genetic Programming , 2017, ACALCI.

[16]  Satish T. S. Bukkapatnam,et al.  Time series forecasting for nonlinear and non-stationary processes: a review and comparative study , 2015 .

[17]  Geoffrey E. Hinton,et al.  Distilling the Knowledge in a Neural Network , 2015, ArXiv.

[18]  Ying Tan,et al.  A multi-resolution-concentration based feature construction approach for spam filtering , 2013, The 2013 International Joint Conference on Neural Networks (IJCNN).

[19]  H. Altay Güvenir,et al.  Voting features based classifier with feature construction and its application to predicting financial distress , 2010, Expert Syst. Appl..

[20]  Wilfried N. Gansterer,et al.  On the Relationship Between Feature Selection and Classification Accuracy , 2008, FSDM.