A number of financial firms have, in recent years, moved towards automating elements of their processes. At the Deep Learning in Finance Summit in London, this March 19 - 20, experts in the space will come together to explore advances in deep learning tools and techniques from the world's leading innovators across industry, research and the financial sector.

Presenting his recent work in recurrent neural networks is Diego Klabjan, a professor at Northwestern University in the Department of Industrial Engineering and Management Sciences. Diego is also Founding Director, Master of Science in Analytics, and the Deep Learning Lab. His expertise is focused on data science and deep learning with a concentration in finance, insurance, and healthcare. Professor Klabjan has led projects with large companies such as The Chicago Mercantile Exchange Group, Intel, General Motors and many other, and he is also assisting numerous start-ups with their analytics needs. He is also a founder of Opex Analytics.

In advance of the summit, Diego has outlined some of the key ways in which his research is progressing and has a real world impact on the financial industry.

Recurrent neural networks are well suited for temporal data. There is abundant work when training data are well defined sequences on the encoding and decoding side in the context of sequence-to-sequence modeling. Consider language-based tasks such as sentiment analysis. A sentence is well defined and maps into the encoder. On the decoding side a single prediction is made (positive, neutral, negative). In financial data, a sequence can go back 10 order book updates, 131 of them, or even 1,000. The same issues is present on the decoding side; do we want to make the prediction for the next one second, one minute, or one hour in increments of 5 minutes?

The length of the input sequence remains a challenging problem and is subject to trial-and-error.

We were able to make advances on how to output only confident predictions in a dynamic fashion. In a very volatile market, a model should be able to reliably make only short term recommendations, while in a stable one, the confidence should increase and more predictions should be made. This is the trait of our new model.

Standard models have a fixed number of layers (think about the number of neurons in each time step). In a challenging market, one should spend more time exploring the patterns and learning while we can only skim and move on in easy times. There is no reason why a model should not follow the same strategy. Another family of models discussed are adaptive computational time that lead naturally to some of the challenges related to time series data. These models dynamically allocate the number of layers in each time and thus the hardness of computation in each time is controlled. First, data scientists do not need to fine tune the number of layers, and, second, the model allocates a lot of time to hard portions of a sequence and just one layer/neuron to easy parts.

All these novel aspects have been tested on a few financial datasets predicting prices of ETLs and commodities. The prediction power is drastically improved by using these enhancements. One vexing challenge remains in evaluation. How does better prediction of prices translate into actual trading and P&L? Stay tuned; to-be-seen.

In Diego’s presentation at the Deep Learning in Finance Summit, the audience will learn about state-of-the-art models and techniques, and he will share a bag-of-tricks to use.
In predicting the movement of prices of correlated securities there are two vexing questions: reliability of predictions and model tuning including the number of layers to use in a deep learning model. We take a deeper dive into how to output only confident predictions in a dynamic fashion, and how to dynamically allocate the number of layers in each time and sequence. The results are discussed on financial market data from an investment firm.