My Questions About Google’s Willow Quantum Chip

When I heard about Google’s Quantum Chip Willow, and it being able to perform tasks in 5 mins which would take a classical computer 10 Septillion years. My mind was filled up with so many questions about this, so I’m gonna try and answer all of those after my research on the same.

NOTE: This post only talks about the QC Willow, and not the basics of quantum computing.

Section 1: Benchmarking

1. How did they exactly test the speed?
   Google assessed the speed of the Willow chip using the Random Circuit Sampling (RCS) benchmark. This benchmark is recognized as one of the most challenging tests for quantum computers, designed to demonstrate their capability to perform tasks that classical computers struggle with.
2. What really happens in RCS?
   - It is quantum computing task where the goal is to generate samples from the probability distribution of outputs produced by a randomly chosen quantum circuit.
   - Random circuit generation: The core idea is to randomly select quantum gates from a set of possible operations and arrange them in a circuit, creating a “random quantum circuit.”
   - Sampling the output: Once the circuit is generated, the task is to repeatedly run the circuit on a quantum computer and collect the resulting measurement outcomes (sampling) from the output probability distribution.
   - Classical comparison: To demonstrate quantum supremacy, the output obtained from the quantum computer is compared to what a classical computer could generate by simulating the same random circuit, which can become computationally expensive for large circuits.

Section 2: Inside the Quantum Chip
Typically, more the qubits (computation units of quantum computers) have a tendency to rapidly exchange information with their environments which makes it difficult to protect the information needed to complete the computation. So, more qubits we use = more errors will occur. But Google’s Quantum Chip willow kind of reversed it. The chip has shown that it can reduce error rates exponentially as more qubits are added, reaching a state known as “below threshold.”

1. How was Willow able to reverse “more qubits = more error”?
   Simply, by arranging the qubits in larger than 3x3 grids like 5x5 and 7x7, which was halving the errors after each increase in the grid size. If they didn’t arrange the qubits like this, the errors probably wouldn’t have gone down. You would think that why wasn’t this tried till now, the answer is that the concept was already present but practical implementation faced a lot of challenges related to complexity, scalability, and technical limitations.
2. What is this quantum error correction really?
   These errors can occur during computation or data transmission, these are due to various sources like — Decoherence (due to the interference from the outside environment), Faulty Gates, Measurement Errors.
   - In QEC, one logical qubit is encoded using multiple physical qubits. This redundancy allows the system to detect and correct errors without losing the encoded information.
   - For example, the Shor code uses nine physical qubits to encode a single logical qubit, while the Steane code uses seven physical qubits.
   - Error Detection : A process called syndrome measurement is utilized where specific measurements are made on the physical qubits to determine if an error has occurred and, if so, which type. This process does not disturb the encoded quantum information but provides information about potential errors.
   - Common types of errors would be — Bit flip errors (from |0⟩ to |1⟩ or vice versa)., Phase flip errors.
   - Error Correction : After detecting the error, an operation is executed to restore the logical qubit to its intended state. For example, if a bit flip is detected on one of the physical qubits, a correction operation like applying a Pauli-X gate is performed to revert it.
   - Threshold Theorem : The effectiveness of QEC relies on the threshold theorem, and it states that as long as the error rate of individual physical qubits is below a certain threshold, it is possible to perform reliable quantum computation using QEC techniques.

Section 3: Practicality and the road ahead

1. Talking about real world applications, at what stage we’re really at? Will this be of any use right now?
   This chip has presented significant advancements in quantum computing (performance and error connection innovation), but its practical applications are still at a very early stage, so no real world use right now.
2. What they really want to achieve from here?
   While the chip has achieved significant milestones, now Google is aiming to apply Willow’s capabilities in fields such as artificial intelligence, drug discovery, and energy systems, but these applications are still in development. Experts, including Hartmut Neven from Google Quantum AI, have indicated that it may take until the end of the decade before a commercially viable quantum chip becomes available. Although, integrating these advancements into broader quantum computing frameworks remains a challenge.
3. Did researchers anticipate the output of this benchmark would be this awesome?
   While they obviously had high expectations, the result exceeded that to a great extent.

References -
Google’s QC Willow blog — https://blog.google/technology/research/google-willow-quantum-chip/

Thanks for reading.