World Business StrategiesServing the Global Financial Community since 2000

Day 1: Wednesday 6th April

08.30 – 09.00

Registration and Welcome Coffee

09.00 - 10.30

Double Session: Alternatives to Deep Neural Networks for Function Approximations in Finance

Vladimir Piterbarg:

MD, Head of Quantitative Analytics and Quantitative Development, NatWest Markets

Alexandre Antonov:

Quantitative Research & Development Lead,
10.30 - 11.00

Morning Break and Networking Opportunities

11.00 - 11.45

New Developments in Deep Pricing

Youssef Elouerkhaoui:

Managing Director, Global Head of Markets Quantitative Analysis, Citi

11.45 - 12.30

Variational Encoder-Generator-Decoder (VEGD) Models for the Interest Rates

Abstract:

  • We propose a variational encoder-generator-decoder (VEGD) model architecture in Q- and P-measure where:
    • Latent space geometry is discovered by pretraining VAE encoder and decoder to optimally represent historical interest rate curves, rather than rate increments
    • Probability distribution over the latent space is determined by the generator located between encoder and decoder
    • Curve and calibration constraints in Q-measure are applied as additional biases of the decoder
  • VEGD model learns the optimal mapping of state variables to latent variables and latent space geometry directly from the data, without committing to an SDE
  • The proposed architecture permits building a wide variety of models with desirable properties depending on the available calibration data, just like with traditional SDE-based models
  • Examples of using VEGD architecture to build machine learning counterparts of short rate models, forward rate models, and curve factor models are provided

Alexander Sokol:

Executive Chairman and Head of Quant Research, CompatibL

12.30 - 13.30

Lunch Break

13.30 - 14.15

“Explainability of Learning Models”

Harsh Prasad:

Principal and CEO, Qxplain
14.15 - 15.00

Forecasting Intraday Stock Returns with Deep Learning Using the Limit Order Book

Petter Kolm:

Clinical Full Professor and Director, Courant Institute of Mathematical Sciences, NYU

15.00 - 15.45

Function approximation in Risk Calculations: When to use Deep Neural Networks and when to use Chebyshev Tensors

Ignacio Ruiz:

MoCaX Intelligence
15.45 - 16.15

Afternoon Break and Networking Opportunities

16.15 - 17.00

Machine Learning for Quant Strategies in Crypto Assets using On-Chain Data

  • On-chain fundamental and flows data for crypto assets
  • Features engineering
  • Generalized fused and group Lasso ML methods for feature selection and model training
  • Efficient solutions for high-dimensional estimation problem
  • Simulation of quant strategies using trained Lasso models

Artur Sepp:

Head Quant, LGT Bank

17.00 - 18.00

Panel: Machine Learning & Quantum Computing

Moderator:

Paul Bilokon:

CEO, Thalesians, Visiting Professor, Imperial College

Blanka Horvath:

Associate Professor in Mathematical and Computational Finance, University of Oxford

Alexander Sokol:

Executive Chairman and Head of Quant Research, CompatibL

Vladyslav Ivanov:

Quantitative Researcher, Outremont Technologies

David Garvin:

Principal Researcher, NEC Australia

 

18.00 - 19.30

Drinks Reception

  • Discount Structure
  • Special Offer
    When two colleagues attend the 3rd goes free!

  • 70% Academic Discount
    (FULL-TIME Students Only)

Event Email Reminder

Error