Using Deep Neural Networks To Predict Foreign Exchange Prices

Oct 15, 2015 · New York, United States of America

In this talk we try to show that neural networks are a useful tool for predicting price changes on the international currency markets. We hope this will highlight the potential of neural networks to solve a large set of the data challenges faced by financial institutions. Specifically we’ll show how we applied recurrent neural networks (RNNs) with long short-term memory (LSTM) using a regularization technique called “dropout” to financial data. We extend work done by Zaremba, Sutskever, and Vinyals in Recurrent Neural Network Regularization.

Speaker Bio:
Stephen Piron is a sometimes-vegetarian, a mediocre computer programmer and life-long friend to the self-starting geek. 
About is a Toronto based fintech startup who believe that artificial intelligence can solve the toughest problems faced by financial institutions. They aim to be the ‘anti-bank’ --- a company where the best and brightest feel comfortable to be technically creative. More info can be found at .

Event organizers

Are you organizing Using Deep Neural Networks To Predict Foreign Exchange Prices?

Claim the event and start manage its content.

I am the organizer

based on 0 reviews