Google today launched Cirq, an open source framework for running algorithms on the quantum computers that will be available in the near future.
A common problem researchers face when designing quantum algorithms for today’s quantum computers – the 50 to 100 qubit Noisy Intermediate-Scale Quantum devices – is in working within the limitations and nuances of the hardware.
Poor mapping between the algorithms and the machines, and ignoring the devices’ complex constraints, inevitably leads to wasted resources and faulty computations.
“The thesis is that when we’re developing algorithms in the NISQ era, you’re going to have to pay attention to hardware. It’s going to become very important. And the algorithm development should aim to deal with those details,” said Dave Bacon, software lead, Google AI Quantum Team.
Announcing the public alpha of the Python-based framework at the First International Workshop on Quantum Software and Quantum Machine Learning at UTS in Sydney, Bacon explained the approach his team had taken to its development.
“The core philosophies we had when we set out to design Cirq: the hardware details need to be part of our programming abstraction. We hear a lot the words ‘hardware agnostic’ used but definitely for NISQ algorithms I just don’t think agnostic is the way to go. I like to say we need to be hardware 'nostic. We need to care about those details,” Bacon said.
“The other approach that we’re trying to take which is slightly different is we think that the hardware…is actually going to be driving the language features, so we are very conservative in our approach to that,” he added.
“We’re going to have data structures, we should have data structures and abstractions that work in different contexts in different ways. Data structures are optimised for writing and compiling these quantum circuits to allow users to get the most out of NISQ architectures,” he said.
Cirq’s features include users being able to “fine tune control over quantum circuits, specify gate behavior using native gates, place these gates appropriately on the device, and schedule the timing of these gates”.
Prior to today’s launch, Google had been working with a number of early adopters including NASA, quantum software company QC Ware, and the UK-based Cambridge Quantum Computing. Bacon called on the audience of global quantum computing and machine learning experts to “shout at him” with additional feedback.
Bundled in with today’s announcement was the release of OpenFermion-Cirq. OpenFermion is a platform for developing quantum algorithms for chemistry problems, and OpenFermion-Cirq is an open source library which compiles quantum simulation algorithms to Cirq.
The Google AI Quantum Team hopes to be able to test Cirq on rival NISQ machines, Bacon said.
“I don’t feel like the community is at a stage where we need to care about that part of the story yet. Other companies may care about that,” he said.
Those rival companies have also released quantum programming languages and platforms over the past year.
Bristlecone to the cloud
Google also announced its intention to make its 72 qubit Bristlecone quantum processor available in the cloud. The company in March said it was “cautiously optimistic” that Bristlecone would achieve ‘quantum supremacy’ the point at which a quantum device can perform a task a classical supercomputer would not be able to complete in a reasonable amount of time.
When Bristlecone is made available to external researchers, Cirq will be the interface in which users write programs for it, the company said.
IBM recently announced a number of commercial partners and start-ups being given to access to its 20 qubit quantum computer via the cloud, among them Sydney quantum technology company Q-Ctrl. Microsoft has a 40 qubit quantum simulator available in its Azure cloud.
Bacon added that his team had an open mind about where Cirq would take them.
Read more: Groupon accuses IBM of patent shake-down
“We’re still evolving at the moment. We’re not sure exactly where this is going to end up. It’s actually been really useful to us to just have an open mind about where we’re going to go,” he said.