HSBC has begun collaborating with Terra Quantum to investigate using hybrid quantum applications to tackle optimisation challenges.
One example of such an issue is collateral optimisation, which is the process of efficiently allocating and managing collateral assets to meet regulatory requirements while minimising costs. This requires mathematical and algorithmic strategies to balance risks, liquidity and profitability.
According to Terra Quantum, existing methods for collateral optimisation primarily rely on linear optimisation solvers, which can sometimes falter when confronted with higher complexities. The company said that the quantum approach could potentially excel beyond traditional methodologies by effectively handling high-dimensional optimisation problems and potentially better scalability.
Vishal Shete, managing director UK at Terra Quantum, said: “Optimisation of capital is one of the core functions in a bank, quantum technologies have the potential to enhance optimisation solutions across many parts of a financial institution, we look forward to realising the benefits of this in the near future.”
HSBC has collaborated with a number of leading technology providers and research laboratories to investigate the potential of applying quantum technologies to real-world problems across the bank.
In 2020, it became a member of thee European NEASQC (Next Applications of Quantum Computing) projects. NEASQC is a consortium of 12 European companies exporing quantum application areas.
The bank’s aim is to be at the forefront of quantum computing, and it is exploring how to integrate quantum computing into its products and services. Along with building a dedicated quantum research team and in-house team of PhD scientists to formalise use cases into deep research projects and develop patents and quantum products, HSBC said it is collaborating across business lines and functions to develop real-world use cases to improve processes to prepare for a quantum-secure economy.
Along with developing a hybrid quantum-classical proof-of-concept to optimise allocation of collateral in the most cost-effective way, the bank is also testing quantum key distribution and pricing optimisation; using quantum computing for random number generation to improve Monte Carlo simulations; and using quantum machine learning to improve fraud detection rates.
Fault-tolerant quantum computing is still some way off, which means applications need to take into account errors that occur in quantum computer hardware. According to McKinsey, before a fault-tolerant quantum computer is available, quantum computing will likely provide speedup for simulations, hybrid machine learning and artificial intelligence, and hybrid optimisation, where classical algorithms can split problems into digestible smaller problems which quantum algorithms can calculate faster.
Markus Pflitsch, CEO and chairman at Terra Quantum, said: “Hybrid quantum algorithms will revolutionise large scale intractable optimisation tasks in the future. We want to demonstrate some of that potential already today.”