It can be frustrating to try to read a book in a noisy room. To focus better, you might put on noise-canceling headphones that detect ambient noise and generate an opposite sound wave to cancel it out. USC professor Daniel Lidar, director of the USC-IBM Quantum Innovation Center, explores a similar approach known as “dynamical decoupling” for protecting fragile quantum information from external disturbances that cause errors.

Just like the noise-canceling headphones, he applies a series of rapid, carefully timed pulses to qubits. These pulses act like a shield, canceling out the effects of external noise and preserving their quantum state for longer. A leading expert in error correction and suppression, Lidar is refining the technique by accounting for the specific properties of the noise affecting the quantum system, allowing for more tailored and efficient ways to pulse them out.

One approach for minimizing quantum errors is to develop better hardware. At USC Dornsife, experimental physicist Eli Levenson-Falk builds qubits out of superconducting circuits and explores new ways to suppress errors caused by the physical environment. His recent research demonstrates an ability to reduce errors by using a qubit as a refrigerator, cooling a circuit and removing photons from it. Removing these photons allows qubits to maintain quantum properties such as superposition and entanglement longer, reducing the likelihood of an error.

Individual qubits are notoriously finicky, but by pairing them with other qubits and organizing them in a specific structure, they become more resistant to decoherence caused by their environment. Building on this idea, a team of USC Dornsife researchers led by physicist Stephan Haas recently demonstrated a significant advancement: By structuring qubits within a carefully designed matrix, the team was able to control and extend the duration these qubits can operate without losing coherence — a breakthrough that could help pave the way for highly reliable quantum sensors and computers.

An appraiser might not know it, but that flaw in your diamond might have big scientific value. USC Dornsife chemist Susumu Takahashi explores the potential for using diamonds as quantum sensors. Small bubbles of nitrogen — often considered impurities — can create vacuums within these gemstones, which are essentially systems closed to the outside world. In fact, they are so resistant to variations in external environments that the nitrogen atoms within these bubbles demonstrate signs of quantum behavior. But what is really special is that these behaviors happen not at temperatures near absolute zero but at temperatures that are normal to our human environment. Takahashi sees promise in using these quantum-enhanced diamonds across a number of applications, including precision medical imaging. Just imagine being able to detect a single cancer cell using quantum diamond technology and removing it before it grows into a tumor.

Think of electrons as tiny spinning tops. In addition to buzzing around an environment to create electricity, they also spin in different ways and can exist in states of quantum superposition. By controlling the spin, scientists can use electrons as qubits that process information. At USC Dornsife, the Ultrafast Quantum Opto-Spintronics Group, led by physicist Kelly Luo, harnesses the power of spin using a unique class of materials that are only a few atoms thick. These ultra-thin materials have special properties that make them perfect playgrounds for electrons — with conditions that allow researchers to observe and test new techniques for manipulating spin. As understanding of this behavior grows, so does the potential for using controlled electrons not only as qubits, but also in quantum sensors that are capable of operating under a variety of conditions.

Imagine using a network of highly sensitive microphones spread across a vast wilderness to listen for the faint sound of a tiny bird. Each microphone captures part of the sound; and by working together, they filter out other noises, making it easier to hear the bird’s call. Collaborating with a multi-university team of researchers, USC engineer and physicist Quntao Zhuang is using a similar approach with a network of quantum-enhanced sensors to hunt for dark matter — a mysterious component of the universe that scientists are confident exists but has never been directly detected. These sensors use quantum entanglement and a special technique known as “squeezing” to reduce background noise, enhancing their ability to detect the elusive signals of dark matter and expand our fundamental understanding of the universe.

Modern biological research generates vast amounts of data — and harnessing this data using machine-learning methods for meaningful insights presents a formidable computational challenge. In a first-of-its-kind study conducted in 2018, a USC Dornsife research team including Rosa Di Felice, Remo Rohs, and Daniel Lidar mapped complex interactions between genes and regulatory proteins. They demonstrated that a quantum computer can achieve an advantage in machine-learning performance over state of the art classical methods using actual biological data. Because many diseases, including cancers, are caused by abnormalities in the way that proteins interact and regulate genes, understanding these relationships could one day underpin personalized treatments and prevention measures.

New materials are waiting to be discovered that could one day point us toward personalized next-gen products like lighter alloys for cars and spacecraft, better batteries, and efficient solar cells. At USC Dornsife, computational physicist Itay Hen is focused on improving the quantum algorithms that could help bring these materials to fruition by simulating how atoms interact. The algorithms crunch numbers in giant tables (matrices), which are then manipulated by multiplying them together many times — a process called matrix exponentiation. Hen devised a clever technique to speed up this process and has applied it to quantum computing, which could eventually enable researchers to simulate materials much more efficiently than before.

Generative artificial intelligence uses machine learning techniques to recognize patterns within vast amounts of data. This training data is then used to generate new information such as text, images, and synthetic datasets to make predictions. While there is enormous potential for quantum systems to greatly enhance artificial intelligence with the ability to analyze even larger datasets, this also comes at a cost. With so much data to navigate, the model can lose its way while training and make mistakes. But USC Dornsife researchers Xiaohui Chen and Quntao Zhuang have developed a clever quantum A.I. learning model that may help avoid this pitfall. It purposely adds errors to the data and trains the model to generate new data from noise by systematically breaking its learning process into smaller, more manageable steps. While functional, quantum A.I. is still some years away, these researchers have built the algorithms and laid out the theoretical basis that may underpin this technology.

Using an IBM quantum computer, USC’s Daniel Lidar and former USC graduate student and IBM researcher Bibek Pokharel have demonstrated a quantum speedup advantage on a mathematical problem they call a “bitstring guessing game.” In this game, a classical computer would require about 33 million guesses to identify a secret 26-bit string, while a perfectly functioning quantum computer could do it in just one guess. By using a quantum algorithm combined with a dynamical decoupling approach to suppress errors, Lidar and Pokharel successfully identified the strings for any number of bits up to 26. Although classical computers can currently solve the problem faster in absolute terms, the study shows that with proper error control, quantum computers can execute algorithms with better time-scaling as the number of bits increases, even in the current era of error-prone quantum computers.

Quantum chemists think big. Like … molecule big! The potential for using complex molecules in quantum information science offers several advantages over atoms or small molecules, including greater versatility and higher information density. But not all molecules are created equal. So how do we know which are best-suited for the job? USC Dornsife quantum chemist Anna I. Krylov develops algorithms and software to screen molecules for suitability as qubits and designs molecular systems for laser cooling and manipulation. The work conducted by her research group is laying a foundation for research that could eventually translate to better energy solutions, better medicine, better materials, and better hardware for quantum information science.