• MyGTC LOGIN
GPU Technology Conference GPU Technology Conference
  • ATTEND
    • Why Attend
    • How to Get There
    • Where to Stay
    • FAQs
    • Pricing
    • Overview
  • AGENDA
    • Session Scheduler
    • Speakers
    • Tracks
    • Training
    • Deep Learning Institute
    • Emerging Companies Summit
  • INCEPTION
  • EXHIBIT
    • Exhibit
    • Floor Plan
    • Expo Map
    • 2017 Partners
  • PRESENT
    • Call for Speakers
    • Talks
    • Instructor-Led Labs
    • Posters
  • RESOURCES
    • GTC On-Demand
    • Media
    • Highlights
    • Poster Gallery
  • CONTACT US
  MUNICH 10-12 OCT 2017
  • REGISTER NOW
  • Finance v
  • Tracks
  • AI and Deep Learning
  • Autonomous Vehicles
  • Professional Visualisation
  • HPC and Supercomputing
  • Virtual Reality and Augmented Reality
  • Autonomous Machines & AI Cities
  • Healthcare
  • Tracks
  • AI and Deep Learning
  • Autonomous Vehicles
  • Professional Visualisation
  • HPC and Supercomputing
  • Virtual Reality and Augmented Reality
  • Autonomous Machines & AI Cities
  • Finance
  • Healthcare

NVIDIA POWERS LATEST ADVANCES IN FINANCE TECHNOLOGIES

GTC brings together some of Europe's top financial institutions to showcase how GPU and deep learning technology are being applied in the Financial Services sector. Sessions will feature a number of user cases including how deep learning solutions can dramatically improve productivity in eCommunications surveillance and enable efficient discovery for GDPR compliance. The Finance track will also highlight current innovation in the traditional Quant analytics application segment.

HEAR FROM INDUSTRY LEADERS IN FINANCE

This track focuses on state-of-the-art research in analytics, speech recognition and fraud detection plus much more. Hear from industry experts, connect directly with NVIDIA engineers, and get valuable, hands-on training from the NVIDIA Deep Learning Institute (DLI). Get the knowledge and experience you need to get the power of deep learning working for you.

Shahzad Chohan

Credit Suisse

Global Head of Future State Engineering and Innovation

Toby Leheup

Credit Suisse

Systems Architect

Ari Juntunen

Elinar

CTO

Pierre Spatz

Murex

Head of Quantitative Research

Steve Karmesin

Numerix LLC

Senior Vice President

Vaivaswatha Nagaraj

Numerix LLC

Senior Software Engineer

James Mesney

KINETICA

Principal Solutions Engineer at Kinetica EMEA

Nigel Cannings

INTELLIGENT VOICE

CTO

Mike Imas

DLI Finance Course Developer

Agenda

NVIDIA Deep Learning Institute

The NVIDIA Deep Learning Institute (DLI) offers hands-on training for developers, data scientists, and researchers looking to solve challenging problems with deep learning.

Join us for the following Finance themed labs on Wednesday, 11th October:

Trading Strategy for Finance using LSTMs
13:00–15:00 h
Level: intermediate

This Lab demonstrates how to structure and train LSTM deep neural networks to predict time series behavior using technical and fundamental inputs. This lab is based on the dataset in the Kaggle contest called the ""Two Sigma Financial Modeling Challenge"", and contains anonymized features pertaining to a time-varying value for financial instruments. We use the LSTM network to train a predictor of the target variable optimized to find the highest possible correlation with the labelled targets. The lab uses the TensorFlow deep learning framework and Python data science tools like Pandas to perform data cleaning, RNN network construction, training, and evaluation.

After taking this lab you will be able to:

  • Structure and train an LSTM network in TensorFlow to accept vector inputs and predict a target
  • Prepare time series data and test network performance using training and test datasets
  • Understand the steps in creating an end-to-end RNN time series prediction algorithm in TensorFlow that could be benchmarked against traditional machine learning techniques
Prerequisites:
  • Working knowledge of basic scientific python
  • Basic level knowledge of TensorFlow

Algorithmic Trading using Deep Autoencoder based Statistical Arbitrage
15:30–17:30 h
Level: intermediate

Note: This lab is taught at GTC Munich as an exclusive beta test offering

Linear techniques such as PCA are the workhorse of creating eigenportfolios that can be used for statistical arbitrage strategies. This lab demonstrates using a deep autoencoder to learn a hierarchical, non-linear basis set that can be used to reconstruct security return data. Anomalous deviations (spreads) in reconstruction error are used as (mean reverting) signals for creating long/short positions. P&L of the strategy can be calculated and benchmarked against traditional linear techniques. This course uses TensorFlow and Python for the code examples covered during the lab session.

After taking this lab you will be able to:

  • Structure and train a deep autoencoder in TensorFlow and Python.
  • Use the autoencoder as an anomaly detector to create an arbitrage strategy and perform hyperparameter optimization over the autoencoder.
  • Calculate P&L of the strategy
Prerequisites:
  • Working knowledge of basic scientific python
  • Basic level knowledge of TensorFlow
  • Knowledge of PCA techniques for statistical arbitrage


 

Shahzad Chohan

CREDIT SUISSE

Global Head of Future State Engineering and Innovation



Shahzad leads the Future State Engineering and Innovation stream in the Semantic Technology, Analytics and Machine Intelligence group. Shahzad led the introduction, build and deployment of the strategic Hadoop project to Credit Suisse, now pioneering the Machine-Learning-on-GPUs space and continues to push technical and innovation boundaries.






 

TOBY LEHEUP

CREDIT SUISSE

Systems Architect



Toby has a degree in Computer and Artificial Intelligence from University of Kent, UK. Toby has pioneered Machine Learning on GPUs at Credit Suisse and introduced the first high-performance Machine Learning stack within the firm. He works extensively with introducing cutting-edge initiatives to solve business problem. Toby has also introduced unique user-friendly tooling for engaging with a Deep Learning platform built within the firm. Previously, Toby has worked on research to detect fraudulent activities within competitive chess and has successfully detected fraudulent agents in large data sets. He has recently managed to go a whole 12 months without crashing his motorcycle, a personal record.






 

ARI JUNTUNEN

Elinar

CTO



Ari Juntunen is co-founder of Elinar Oy Ltd. He has a long (20+ years) history on content centric processes and has been heavily involved with Elinar.AI development for past two years. Ari has been providing international consultancy services in Northern Europe on Records Management, Text Analytics and AI. He is a passionate innovator in constant search for technology that can make daily lives of business users more rewarding by eliminating menial tasks from their daily workload using AI. And naturally to get rid of unnecessary workforce.






 

PIERRE SPATZ

Murex

Head of Quantitative Research



Pierre Spatz heads the quantitative analysis team of Murex, a world leader in trading and risk management software. He holds a master's degree in computer engineering and applied mathematics from ENSIMAG in Grenoble, France.






 

STEVE KARMESIN

Numerix LLC

Senior Vice President



Steve Karmesin received his Ph.D. in Applied and Computational Mathematics from Princeton University in 1991. After working at Caltech/JPL he went to the Advanced Computing Lab at LANL where he worked on code architecture techniques massively parallel computation and led the POOMA team. Since 2005 he has worked at Numerix LLC, and is presently Senior Vice President and head of the CrossAsset team. He is based in Santa Fe, New Mexico.






 

VAIVASWATHA NAGARAJ

Numerix LLC

Senior Software Engineer



Vaivaswatha Nagaraj received his Masters degree in Computer Science from Indian Institute of Science, Bangalore in 2014. He has worked as a compiler engineer at NVIDIA, contributing to the Tegra GPU compiler. He has worked on the LLVM compiler while employed at Compiler Tree Technologies. He has worked at Numerix since 2016 and is currently working on dynamic compilation and other techniques to improve performance of Monte Carlo simulations used in derivative pricing.






 

James Mesney

KINETICA

Principal Solutions Engineer at Kinetica EMEA



James is a Principal Solutions Engineer for Kinetica. James has 25 years of experience working with numerous vendors in the analytics, BI, data warehousing, and the Hadoop spaces. James has a degree in Computer Science from Staffordshire University, and lives in Hampshire with his wife and two daughters. He enjoys travel and is a keen skier, terrible boarder, cook, runner, photographer and cyclist.






 

Nigel CANNINGS

INTELLIGENT VOICE

CTO



Nigel qualified as a solicitor in 1993 and has worked for some of the world's largest law firms and software companies. Nigel has been the CTO of Intelligent Voice, Speech Recognition Experts, for the last 7 years. As a keen technologist, Nigel is always on the lookout for new challenges and so is often seen finding new ways of stretching existing techniques and technology. This has led to interesting commissions such as recovering data from an "unbreakable" portable recording device and he has gained UK government recognition in the form of a sizable grant, for Intelligent Voice to explore leading edge problems in speech research such as ultra-high speed GPU accelerated speech recognition and emotional analysis of telephone calls. Nigel is regularly invited to speak at various conferences worldwide, including the NVIDIA GPU Technology Conference in San Jose California in April 2016, where he outlined his idea to get a deep Convolution Neural Network (CNN) to perform speech recognition. This work resulted in Intelligent Voice being awarded for innovation for their SmartTranscript™ and winning almost $100,000 in prizes. Additionally Nigel was invited to speak at GTC Europe 2016 in September, presenting on Intelligent Voice’s revolutionary work on Credibility Analysis in the GPU Age; Rutgers’ 38th World Continuous Auditing and Reporting Symposium in the US in November; the Deep Learning Fintech Event in March 2017 and most recently Nigel presented a talk at the 2017 NVIDIA GPU Technology Conference in San Jose California, speaking on the subject of Encrypted Deep Learning – A Guide to Privacy Preserving Speech Processing”.






 

Mike Imas

 

DLI Finance Course Developer



Mike has worked in finance for over 20 years and has extensive experience working with Statistical Arbitrage, Quantitative Finance, Algorithmic Trading and a wide range of Finance technologies. He has worked for many global investment banks and hedge funds in New York and London. Mike has extensive experience in applying data science in general and deep learning in particular to various fields of Finance. He holds masters degrees in Theoretical Physics, Computer Science and an MBA.





< >

#GTC17EU

 

SUBSCRIBE FOR UPDATES

GO
©2017 NVIDIA Corporation. All rights reserved
GPU Technology Conference   Legal   Cookie Policy   Privacy
GPU Technology Conference
MUNICH 10-12 OCT 2017
  • ATTEND v
    • Why Attend
    • How to Get There
    • Where to Stay
    • FAQs
    • Pricing
    • Overview
  • AGENDA v
    • Session Scheduler
    • Speakers
    • Tracks
    • Training
    • DLI and Hands-on Labs
    • Emerging Companies Summit
  • PRESENT v
    • Call for Speakers
    • Talks
    • Instructor-Led Labs
    • Posters
  • INCEPTION
  • EXHIBIT v
    • Exhibit
    • Floor Plan
    • Expo Map
    • 2017 Partners
  • RESOURCES v
    • GTC On-Demand
    • Media
    • Highlights
    • Poster Gallery
  • CONTACT US
  • REGISTER
  • LOGIN 

#GTC17EU

 

SUBSCRIBE FOR UPDATES

GO
©2017 NVIDIA Corporation. All rights reserved
GPU Technology Conference   Legal   Cookie Policy   Privacy
Subscribe for updates
Submit

*Mandatory