Videos

Riskfuel is focused on providing accurate and inexpensive generation of real-time valuations and risk sensitivities for financial instruments. In this talk, Ryan will highlight some of Riskfuel’s recent achievements including the fast and accurate deep neural network (DNN) representation of a high dimensional Bermudan swaption model. He will also discuss some novel techniques recently developed at Riskfuel, including a technique to handle the discontinuities present in the valuation functions of many common financial instruments.

This talk is an introduction to generative modelling, including a walkthrough of one of the most utilised generative deep learning models – the Variational Autoencoder (VAE). We will also explore examples of state-of-the-art output from Generative Adversarial Networks (GANs) and Transformer based architectures and see how generative models can be used in a reinforcement learning setting (World Models paper) The talk will conclude with why I believe generative models will play a crucial part in the quest to build Artificial General Intelligence (AGI).

One of the key factors for understanding exchange rates is that of capital flows. If a country experiences large capital inflows it helps to support its currency, whilst large outflows will weaken it. In order to forecast currency moves, we need to be able to forecast these underlying capital flows in a timely way. In this webinar, Jens will discuss how capital flow data can be used to forecast the dollar, in particular using high frequency estimates of these capital flows from Exante Data.

The talk studies a set of intertwined information-asymmetric game theoretic models of viral mimicry, immune evasion, asymptomatic selective sweeps, crisis economy and… moves beyond. It will highlight how quickly this ‘wicked’ problem has led to deceptive and shifting Nash equilibria but without an exit strategy in sight. In the absence of clarity (e.g., access to complete information) and yet facing in Covid a capricious and complex conspirator, we overview an exemplary solution, created by RxCovea, and examine how it might help.

QuantLib is an open-source library for quantitative finance. Now in its twentieth year, it is known and used by numerous practitioners and firms. However, it might be difficult to start using the library, since its architecture needs to be understood in order to use it to one’s advantage. In this talk, I will describe and motivate the core design of QuantLib through a few live examples of its usage.

The problem at hand is partly the application of software engineering best practices to AI, but more so the evolution of software engineering to attend to software-intensive systems that contain AI components. In this lecture, Grady Booch will examine both dimensions: emerging AI architectures, neuro-symbolic systems, designing/testing/deploying/refactoring/maintaining systems with AI components; the future of software engineering.

This talk introduces Q-learning as a successful algorithm in reinforcement learning. It illustrates the application of a DQL agent to a game from the OpenAI Gym environment. It also illustrates how the same DQL agent can learn to trade financial instruments. The examples are based on self-contained Python code. It is also shown how such a trading bot can be easily deployed to trade algorithmically intraday based on The AI Machine (http://aimachine.io).

We show how reinforcement learning can be used to derive optimal hedging strategies for derivatives when there are transaction costs. We illustrate our approach by showing the difference between using delta hedging and optimal hedging for a short position in a call option when the objective is to minimize a function equal to the mean hedging cost plus a constant times the standard deviation of the hedging cost. Two situations are considered. In the first, the asset price follows a geometric Brownian motion. In the second, the asset price follows a stochastic volatility process. The basic reinforcement learning approach is extended in a number of ways.

The first and only book to systematically address methodologies and processes of leveraging non-traditional information sources in the context of investing and risk management Harnessing non-traditional data sources to generate alpha, analyze markets, and forecast risk is a subject of intense interest for financial professionals. A growing number of regularly-held conferences on alternative data are being established, complemented by an upsurge in new papers on the subject. Alternative data is starting to be steadily incorporated by conventional institutional investors and risk managers throughout the financial world. Methodologies to analyze and extract value from alternative data, guidance on how to source data and integrate data flows within existing systems is currently not treated in literature. Filling this significant gap in knowledge, The Book of Alternative Data is the first and only book to offer a coherent, systematic treatment of the subject.

Tail hedging does not necessarily involve options and the annoying time decay of premium. Dr. Chan will discuss how his fund utilizes a simple intraday trend-following strategy called Tail Reaper to generate enormous alpha during market crises, with a large dose of machine learning.