## NeuralEF: Learning the Efficient Frontier

*Sept 2023 — Accepted into NeurIPS2023 — *We show how neural networks can be used to solve discontinuous convex optimization problems with heterogeneous linear constraints, which are used in pricing basket option derivatives, by reformulating them as a sequence SEQ2SEQ problems.

##### Volatility Surface Completion

## FuNVol: A Multi-Asset Implied Volatility Market Simulator

*March 2023 — A* new approach for generating sequences of implied volatility (IV) surfaces across multiple assets that is faithful to historical prices.

##### Model Acceleration

## Looking for Trouble: Validating ML Pricers

*Nov 2021 — *Learn about how we validate our super-fast models and prove that we achieve performance improvements of more than a million-fold without compromising accuracy.

##### Volatility Surface Completion

## Stop Messing with Your Volatility Surface

*October 2021 — *Riskfuel has developed a technology to automatically complete volatility surfaces for even the most illiquid assets.

##### Model Acceleration

## Libor Prompts Quantile Leap: Machine Learning for Quantile Derivatives

*July 2021 — *We show how deep neural networks can be trained to quickly and accurately calculate the value of exotic quantile derivatives.

##### Volatility Surface Completion

## Arbitrage-Free Implied Vol Surface Generation with Variational Autoencoders

*Aug 2021 — *We propose a hybrid method for generating arbitrage-free implied volatility surfaces consistent with historical data by combining model-free Variational Autoencoders with continuous time stochastic differential equation driven models.

##### Volatility Surface Completion

## Hands-Off Approach to Completing Implied Volatility Surfaces

March 2021 — In this talk, Riskfuel’s Director of R&D explains how variational autoencoders can remove human bias from this procedure and let the data speak for itself through unsupervised learning.

##### Volatility Surface Completion

## Variational Autoencoders: A Hands-Off Approach to Volatility

*Feb 2021 — *Variational autoencoders can be used to construct a complete volatility surface when only a small number of points are available without making assumptions about the process driving the underlying asset or the shape of the surface.

##### Model Acceleration

## Ultra-fast and Accurate Derivatives Pricing with Deep Learning

*July 2021 — *We show how deep neural networks can be trained to quickly and accurately calculate the value of exotic quantile derivatives.

##### Model Acceleration

## Exploring Riskfuel's Bermudan Swaption Pricing Demo

*Feb 2020 — *This article describes the Bermudan Swaption and its valuation models, and discusses a small case study to illustrate the accuracy of the Riskfuel model and compare its run-time performance again the target Quantlib model.

##### Model Acceleration

## 1,000,000x faster models: how it works

*Jan 2020 — *Deep neural networks can be trained to learn a functional approximation of derivatives valuation models that use mathematic simulations. Riskfuel AI-based technology cuts the computation costs to virtually zero allowing for on-demand recalculation of portfolio values and a complete up-to-the-second view on risk

##### Model Acceleration

## Deeply Learning Derivatives: The Paper that Started It All

*Oct 2018 — *This paper shows how we can use deep learning neural networks to value derivatives. The approach is broadly applicable, and we use a call option on a basket of stocks as an example. We show that the deep learning model is accurate and very fast, capable of producing valuations a million times faster than traditional models.

**Riskfuel**

**1,000,000x**