Nov 22, 2021

By Ian Gormely

A start-up cofounded by Vector Institute postgraduate affiliate Estelle Inack is the first company to be spun-out from Vector. yiyaniQ, uses a neural network to get faster and more accurate derivative pricing calculations, one of finance’s most difficult problems.

The quantum technology company was born out of Inack’s research, which sits at the intersection of quantum physics and AI. Though she lacked significant business experience, “I am very motivated to prove the statements that some of us academics make in the conclusions of our articles, that the techniques we are developing could solve real world problems.”

She and her co-authors, Vector graduate students Mohamed Hibat-Allah and Roeland Wiersema, Vector Faculty Affiliate Roger G. Melko, and Vector Faculty Member Juan Carrasquilla, showcased the neural network they developed in a paper recently published by Nature Machine Intelligence. It details how they were able to simulate both thermal and quantum annealing – a process that can help pick the best solution when problems are complex, and many solutions are possible – simultaneously.

After patenting the technology with help and advice from Vector’s Commercialization team Inack, who is also the Francis Kofi Allotey Fellow at Perimeter Institute, where she works as part of the Perimeter Institute Quantum Intelligence Lab, was faced with a choice: license it to a third party or commercialize it herself. Explaining why she chose the latter, she says “I felt that I had a better grasp of the model’s capabilities.” The company’s name reflects the personal connection to its co-founder. It comes from Basa’a, a local language used in her native Cameroon. “Yi means intelligence, and yaani means tomorrow,” she says. “The Q is for quantum, of course.”

To make up for her relative lack of business experience, Inack brought on Behnam Javanparast, a physicist with experience in the financial industry, as co-founder and CEO. They met at the Creative Destruction Lab (CDL) in July where Inack was participating in an accelerator program. “We had a few initial chats about hard problems in capital markets derivative trading and how we might be able to speed up and gain accuracy using AI and quantum technologies,” says Javanparast. “We started working on initial algorithms to tackle toy problems right away.

They credit Vector for support and advice throughout the journey from research to reality.

“Having access to Vector’s compute resources for the project was crucial,” says Inack. Through Vector’s commercialization team they were able to gain access to educational resources around intellectual property and commercialization, and contacts at incubators including NextAI, Entrepreneur First and CDL. “Right now, they are helping us with introductions to some financial institutions which is essential for us.”

“It’s been great to have Vector as yiyaniQ’s partner,” adds Javanparast. “We’re looking forward to future collaborations.”

Inack and Javanparast haven’t decided whether to take the company’s services deeper into finance or broaden yiyaniQ’s scope into other industries. Finding the most energy efficient way of folding a protein, the best schedule for power generation, or the most efficient layout for an integrated circuit are all optimization questions. As yiyaniQ’s only employees (Vector Faculty Member Carrasquilla, who is also a Perimeter Visiting Fellow, is an adviser), they have limited capacity. “We’re testing both our market hypothesis and our technological hypothesis,” says Inack. “We are conducting this research to validate the commercial need for this. It’s a busy time, but it’s great.”