Augmented Bilinear Network for Incremental Multi-Stock Time-Series Classification

23 Pages Posted: 20 Jan 2023

See all articles by Mostafa Shabani

Mostafa Shabani

Aarhus University

Dat Thanh Tran

Tampere University

Juho Kanniainen

Tampere University

Alexandros Iosifidis

Aarhus University - Department of Engineering

Abstract

Deep Learning models have become dominant in tackling financial time-series analysis problems, overturning conventional machine learning and statistical methods. Most often, a model trained for one market or security cannot be directly applied to another market or security due to differences inherent in the market conditions. In addition, as the market evolves over time, it is necessary to update the existing models or train new ones when new data is made available. This scenario, which is inherent in most financial forecasting applications, naturally raises the following research question: How to efficiently adapt a pre-trained model to a new set of data while retaining performance on the old data, especially when the old data is not accessible? In this paper, we propose a method to efficiently retain the knowledge available in a neural network pre-trained on a set of securities and adapt it to achieve high performance in new ones. In our method, the prior knowledge encoded in a pre-trained neural network is maintained by keeping existing connections fixed, and this knowledge is adjusted for the new securities by a set of augmented connections, which are optimized using the new data. The auxiliary connections are constrained to be of low rank. This not only allows us to rapidly optimize for the new task but also reduces the storage and run-time complexity during the deployment phase. The efficiency of our approach is empirically validated in the stock mid-price movement prediction problem using a large-scale limit order book dataset. Experimental results show that our approach enhances prediction performance as well as reduces the overall number of network parameters.

Keywords: Deep Learning, Low Rank tensor decomposition, Limit Order Book data, Financial time-series analysis

Suggested Citation

Shabani, Mostafa and Thanh Tran, Dat and Kanniainen, Juho and Iosifidis, Alexandros, Augmented Bilinear Network for Incremental Multi-Stock Time-Series Classification. Available at SSRN: https://ssrn.com/abstract=4332126 or http://dx.doi.org/10.2139/ssrn.4332126

Mostafa Shabani (Contact Author)

Aarhus University ( email )

Dat Thanh Tran

Tampere University ( email )

Tampere, FIN-33101
Finland

Juho Kanniainen

Tampere University ( email )

P.O. 541, Korkeakoulunkatu 8 (Festia building)
Tampere, FI-33101
Finland

HOME PAGE: http://https://sites.google.com/site/juhokanniainen/

Alexandros Iosifidis

Aarhus University - Department of Engineering ( email )

Inge Lehmanns Gade 10
Aarhus C, 8000
Denmark

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
15
Abstract Views
103
PlumX Metrics