
Podcast: Dario Villani on managing a hedge fund with machine learning
Duality’s CEO discusses key to machine learning success, and the influence of Renaissance’s Jim Simons

In this episode of Quantcast, Risk.net speaks with Dario Villani, co-founder and chief executive officer of Duality Group, a New York-based hedge fund, and co-winner of our inaugural Buy-side quant of the year award in 2016.
Duality uses machine learning-led algorithms to trade US stocks, exchange-traded funds and global futures.
Villani is a machine learning evangelist. He says it beats any traditional model for capturing the structure of complex systems, of which financial markets are an example, and believes Duality is one of the very few investment firms that uses it not just for data manipulation, trade execution or optimisation, but also for forecasting.
Villani discusses Duality’s use of machine learning, explains his against-the-tide views on interpretability and overfitting, and shares the lessons he learned from Jim Simons, co-founder of legendary quant hedge fund Renaissance Technologies.
Index
00:00 Introduction
02:03 Duality’s investment strategy and why it uses machine learning
06:50 Data proliferation
11:02 How Duality uses machine learning
15:33 Interpretability and overfitting
29:40 How to spot flawed ML strategies
36:46 Mean field games
42:00 Operational challenges and talent acquisition
52:05 Lessons from Jim Simons
54:22 Physics and finance
To hear the full interview, listen in the player above, or download. Future podcasts in our Quantcast series will be uploaded to Risk.net. You can also visit the main page here to access all tracks, or go to the iTunes store or Google Podcasts to listen and subscribe.
Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.
To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe
You are currently unable to print this content. Please contact info@risk.net to find out more.
You are currently unable to copy this content. Please contact info@risk.net to find out more.
Copyright Infopro Digital Limited. All rights reserved.
As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (point 2.4), printing is limited to a single copy.
If you would like to purchase additional rights please email info@risk.net
Copyright Infopro Digital Limited. All rights reserved.
You may share this content using our article tools. As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (clause 2.4), an Authorised User may only make one copy of the materials for their own personal use. You must also comply with the restrictions in clause 2.5.
If you would like to purchase additional rights please email info@risk.net
More on Cutting Edge
Quantum path integrals for default intensity models
A method to price credit derivatives via default intensity approximation is presented
Podcast: Villani and Musaelian on a quantum boost for machine learning
Classical methods struggle with highly dimensional problems. Quantum cognition takes a different approach, as hedge fund duo explain
Simulation of Heston made simple
A new way to apply the classic stochastic volatility model is presented
Supervised similarity for high-yield bonds
Quantum cognition ML is used to identify tradable alternatives for high-yield corporate bonds
Podcast: Fabrizio Anfuso on computing for Archegos-like event exposures
BoE quant discusses top-down counterparty risk framework using Gaussian distributions and copulae
Gaussian GenAI: synthetic market data generation
A method to generate financial time series with mixture models is presented
The WWR in the tail: a Monte Carlo framework for CCR stress testing
A methodology to compute stressed exposures based on a Gaussian copula and mixture distributions is introduced
Estimating mean reversions in interest rate models
The speed of factors’ mean reversion in rate models is estimated