Quantum Innovators 2026

Alexander Anferov, ETH Zurich

Hybrid systems such as microwave–acoustic platforms are particularly attractive for quantum experiments since they can enable single-photon interactions between many mechanical modes. Since these multicomponent quantum systems are limited by their collective quantum coherence, it is vital to understand and optimize the coherence of each subsystem.

In this talk I will first discuss how lessons learned from developing superconducting qubits at high-frequencies (and high-temperatures) can be useful for optimizing microwave qubit performance in the demanding context of a hybrid experiment. Shifting focus to the acoustic subsystem, I will discuss how a combination of classical and quantum experiments can build an understanding the composition of high-overtone bulk acoustic resonators, highlighting how improved mechanical performance achieved high quantum coherence cooperativity. Lastly, I will discuss our ongoing research into practical applications of this hybrid microwave-acoustic platform for quantum simulation, and quantum detection.

During my undergraduate studies at Caltech, I investigated Josephson parametric amplifiers in Prof. Oskar Painter's lab. Afterwards, I transitioned to a new frequency range, studying millimeter-wave superconducting quantum devices with Prof. David Schuster at the University of Chicago. After completing my PhD, I spent a year exploring microwave qubit fabrication techniques with Prof. Andrew Cleland. Now looking to branch out into different flavors of quantum systems, I joined the group of Prof. Yiwen Chu at ETH Zurich to explore applications of quantum acoustic resonators.

Krishna Coimbatore Balram, University of Bristol

Piezoelectric devices underpin modern smartphones by enabling compact, high performance RF filters. This demand in turn has rapidly advanced the development of new materials and devices geometries that can efficiently manipulate GHz frequency acoustic waves. Here I will discuss how these developments can be harnessed in new directions in microwave signal processing and quantum devices with a focus on two key themes: engineering wavelength scale confinement of GHz frequency acoustic waves and building efficient microwave interfaces to spins and light.

Krishna C Balram is currently a professor of nanoscale device engineering at the University of Bristol. His primary interest is the development of nanoscale devices with efficient interfaces between optical, electrical and mechanical degrees of freedom for applications in classical and quantum information processing and sensing. His work has been recognized by the award of starting and consolidator grant awards from the European Research Council.

Oana Bazavan, QuEra Computing Inc.

Parallel entangling operations across a large neutral-atom register enable transversal gates and fault-tolerant primitives that make these platforms attractive for scalable computation. The number of atoms that can be entangled in a single parallel operation therefore sets a hard limit on circuit depth and atom-transport overhead. Extending it has proven difficult: maintaining uniform, high-intensity excitation over a growing area places demands on available laser power, and gate fidelity degrades as the addressed region grows. I will begin by presenting progress at QuEra Computing on high-fidelity entangling gates on large numbers of atoms.

If time allows, I will then turn to my PhD work on interactions in continuous-variable systems. Higher-order nonlinear interactions on a harmonic oscillator produce non-Gaussian states, which are a resource for real-time simulation of many-body models. However, these interactions typically become exponentially weaker as their order increases. Using the motion of a trapped ion coupled to its spin, we built higher-order interactions from spin-dependent linear ones alone and reached far higher strengths than conventional methods allow. This led to the first realisation of fourth-order squeezing ("quadsqueezing"), at rates more than 100 times faster than previously possible. The approach imposes no fundamental limit on interaction order and extends to any platform with spin-dependent linear bosonic interactions.

I am a quantum scientist at QuEra Computing, working on entangling gates for large-scale neutral-atom quantum computers. I completed my PhD in 2024 at the University of Oxford under David Lucas, focusing on entangling gates and continuous-variable operations in trapped ions, and stayed on as a postdoctoral researcher to apply them to quantum simulation of lattice gauge theories in hybrid qubit-oscillator systems. I hold an MPhys from the University of Manchester (2019). My expertise spans experimental quantum computing with both trapped-ion and neutral-atom platforms.

Kate Fenwick, National Research Council of Canada (NRC) 

Quantum computing offers a promising pathway to solving problems that are intractable for classical computers. Despite impressive progress, the development of a scalable quantum information processor remains an open challenge. Several hardware platforms remain contenders, with photonic systems being a strong candidate. Until recently, state-of-the-art photonic quantum processors have used path encoding, which often requires an exponentially increasing number of optical components, making these schemes unavoidably large or lossy in some cases. Time-bin encoding has emerged as a promising method to reduce the number of optical components in a photonic quantum processor by enabling operation along a single optical path. Yet, a barrier to the widespread adoption of time-bin encoding is the challenge of maintaining phase stability across the entire device for extended periods of time. We overcome this obstacle here through ultrafast time-bin encoding, in which time bins are separated by just a few picoseconds. This talk will present our ability to generate, manipulate, and measure these ultrafast time bins at the single-photon level, demonstrating the promise of our platform across a wide range of applications.

Kate is a Research Associate, holding a Luise and Gerhard Herzberg Fellowship, in the Ultrafast Quantum Photonics group at the National Research Council of Canada (NRC). She completed her PhD studies in physics at the University of Ottawa, where she held a Vanier Scholarship. Prior to her PhD, she completed her BSc and MSc at Queen's University, also in physics. Kate's primary research interests lie in the field of ultrafast quantum photonic technologies, with a focus on ultrafast photonic quantum information processing.

Sasha Geim, Harvard University

Neutral-atom quantum processors have emerged as a versatile platform for quantum science, enabling fully programmable, high-fidelity control of hundreds to thousands of atomic qubits in parallel. Here, we first leverage these capabilities to engineer and study many-body dynamics in a Rydberg atom array, combining efficient analog many-body evolution, programmable digital measurements, and loss-based error mitigation. As an example, we prepare an out-of-equilibrium Rokhsar–Kivelson-type quantum spin liquid and observe its characteristic features, including the absence of local order and extended many-body coherence. We further measure correlations in good agreement with field-theory predictions and experimentally explore finite-size effects. Second, we pursue a complementary approach based on gate-based simulation and computation, demonstrating two-qubit entangling operations with raw fidelities of 99.84% as a key building block for quantum circuits. Together, these results establish a foundation for programmable control of complex quantum systems, with applications ranging from fundamental many-body physics to error-corrected quantum information processing.

Sasha is a PhD student in experimental physics in the group of Prof. Mikhail Lukin at Harvard University. Her research focuses on quantum computation and simulation using Rydberg atom arrays, ranging from practical implementations of fault-tolerant computation to analog quantum simulation of condensed matter systems. She is particularly interested in developing new experimental approaches that leverage the programmability of the neutral atom platform to enable precise quantum information processing.

Connor Holland, Stanford University

Molecular tweezer arrays have emerged as a powerful new platform for quantum science, combining rich molecular structure with single-particle detection and control capabilities. In this talk, I will discuss my work to bring CaF molecules under full quantum control, including control over the interactions between molecules. These advances have enabled studies of elementary excitations in many-body dipolar spin models, which are prototypical models describing quantum magnetism. I will share how these same spin models can generate spin squeezed states, which harness many-body entanglement to surpass classical limits on sensor precision. These entangled states feature bi-partite entanglement and EPR steering, and can be transferred to a non-interacting encoding, where quantum-enhanced metrological advantage persists for over 100ms. These investigations open opportunities to employ molecular qudit encodings for quantum simulation and computation, generate robust and scalable spin squeezing in molecule-based sensors, and explore chemical reaction dynamics at cold and ultracold temperatures.

Connor Holland is a postdoctoral scholar in the Simon Lab at Stanford University. His research focuses on quantum networking and quantum simulation using neutral atom optical cavity arrays. He earned his bachelor’s degree from Stanford in 2018, and his PhD from Princeton in 2026 under Lawrence Cheuk. Connor’s work drove major advances in molecular qubit quantum control and dipolar spin model simulation using molecular tweezer arrays.

Holly Stemp, Massachusetts Institute of Technology

Gate-defined quantum dots represent a promising candidate for a scalable qubit platform. A key advantage of quantum dots is their small physical footprint, which could enable the integration of many millions of qubits on a single chip. However, this high qubit density creates challenges in routing the on-chip classical control signals needed to scale these systems to a size capable of solving problems of real-world relevance. To address this, long-range spin coupling mechanisms are needed to connect spatially sparse arrays of spin qubits. We propose a novel coupling scheme in which a superconducting qubit mediates interactions between distant quantum dot spin qubits. To implement this approach, we have developed a hybrid semiconductor-superconductor measurement architecture, drawing on established engineering practices from the superconducting qubit community. In this talk, I will present characterization measurements of both the superconducting and quantum dot qubits within this hybrid platform. I will also discuss our progress toward 3D integration of the two qubit types via flip-chip bonding, a key milestone toward realizing hybrid quantum devices.

Holly carried out a master’s degree in physics at the University of Surrey in the UK. As part of her master’s degree she spent 10 months working at Oak Ridge National Laboratory in Tennessee, USA in both the department of nuclear astrophysics and quantum computing. She completed her PhD in the group of Prof. Andrea Morello at the University of New South Wales in Sydney, Australia, where her project consisted of characterizing multi-qubit operations in a system of two exchange-coupled donor spin qubits. In her role as a postdoc in the Engineering Quantum Systems group, led by Prof. William oliver at MIT, Holly is focused on designing and implementing hybrid spin/superconducting qubit architectures.

Yujie Zhang, Institute for Quantum Computing, University of Waterloo

Entanglement certification commonly relies on quantum state tomography with trusted measurement calibration or on Bell tests that detect only a subset of entangled states. In this talk, I will present a self-consistent approach to entanglement certification based on generalized noncontextuality. Assuming tomographic completeness, the method uses experimentally inferred operational identities to certify entanglement without prior measurement calibration. Its conclusions are independent of tomographic gauge freedom, and it can, in principle, certify any entangled state. A photonic demonstration of this approach will be presented, certifying entanglement in states beyond the reach of Bell and steering inequalities. Finally, I will introduce a unified notion of classicality for individual quantum processes and outline a concrete route toward self-consistent certification of quantum nonclassicality beyond entanglement.

Yujie Zhang received his PhD in Physics from the University of Illinois at Urbana-Champaign in 2023 and his BSc in Physics from Nanjing University in 2017. He is a quantum information theorist and postdoctoral fellow at the Institute for Quantum Computing, University of Waterloo. His research lies at the intersection of quantum foundations, quantum information theory, and quantum optics, drawing on experience in both theory and experiment. He studies how genuinely quantum phenomena can be understood, certified, and harnessed for quantum information processing. Looking ahead, he aims to use insights from quantum foundations to clarify the origins of quantum advantage in existing protocols and develop new approaches to quantum sensing, networking, and computation.

Antonio Anna Mele, Freie Universität Berlin

Quantum process tomography asks how many uses of an unknown quantum channel are needed to learn a classical description that is accurate in diamond distance. While the analogous problem of quantum state tomography has been largely settled over the past decades, the corresponding question for general quantum channels remained open beyond special cases.

In this talk, I will show that a generic d-dimensional channel can be learned using O(d⁴/ε²) channel uses, and more generally that O(d_in d_out k/ε²) uses suffice for channels of Kraus rank k, with optimal dependence on the dimension parameters. Perhaps surprisingly, the natural strategy of performing tomography on the Choi state already achieves this scaling. The main challenge is to avoid the dimension loss arising from a naive conversion between trace distance on Choi states and diamond distance on channels. I will explain how random purification and a direct diamond-distance analysis overcome this obstruction. Finally, I will briefly discuss recent work showing that quantum memory can provide a provable advantage for quantum process tomography.

Based on:
- Optimal learning of quantum channels in diamond distance: https://arxiv.org/abs/2512.10214
- Quantum memory advantage for quantum process tomography: https://arxiv.org/abs/2607.13476

Antonio Anna Mele is a PhD researcher in quantum information at FU Berlin and a 2025 Google PhD Fellow in Quantum Computing. His research has focused on quantum learning theory--how can we efficiently extract useful information, or learn a compact classical model, from a complex quantum system?--as well as on the impact of noise on quantum circuits and their classical simulability. His work has also uncovered mathematical structures underlying quantum systems, from Gaussian and non-Gaussian bosonic and fermionic systems to Clifford symmetries and random unitaries. Looking ahead, he is increasingly interested in quantum algorithms and quantum error correction, with the goal of pushing quantum computing toward genuinely useful applications, particularly as increasingly capable AI models enable previously unimaginable ways of doing research and accelerate the search for such applications.

Francesco Anna Mele, Caltech

The talk will be based on our recent works at the intersection of two important fields of quantum information: quantum learning theory and continuous-variable (CV) systems. Quantum learning theory addresses the question of how to extract classical information from quantum systems as efficiently as possible. CV systems are ubiquitous in nature and in quantum technologies, as they model bosonic systems and quantum optical platforms. The intersection of these two fields raises many interesting questions, some of which are addressed in our recent works. The first natural question is: what is the ultimate achievable performance of tomography for CV systems? We answer this question by establishing the optimal sample complexity of tomography of Gaussian states (an efficient task) and non-Gaussian states (an extremely inefficient task). Other natural questions explored in our recent works include: How does the sample complexity of CV tomography grow with the degree of non-Gaussianity of the unknown state? How can we efficiently learn Gaussian processes? And how can we efficiently test whether an unknown CV state is Gaussian or far from the set of Gaussian states? As a by-product of our analysis, we establish mathematical tools that may be of independent interest, including (i) bounds on the trace distance between CV states in terms of their covariance matrices, and (ii) a Gaussian version of the recently introduced random purification channel.

Francesco Anna Mele was born in Italy in 1997. He received the B.Sc. and M.Sc. degrees in Physics from the University of Pisa, Italy, and an additional degree in Physics from Scuola Normale Superiore (SNS), a special-status university for advanced studies in Pisa, Italy, in 2021. He is about to defend his Ph.D. in Nanoscience at SNS, advised by Vittorio Giovannetti and Ludovico Lami. He was a Student Researcher at Google Quantum AI during the summer of 2026, working with Tom O'Brien. He will join the California Institute of Technology in September 2026 as an IQIM Postdoctoral Scholar. His research interests include all aspects of quantum information and computation.

Joseph Carolan, University of Maryland, College Park

In the welded tree problem, an algorithm is tasked with navigating a graph formed from
two binary trees joined at the leaves through a “weld” of connecting edges. A quantum walk
can navigate from root to root exponentially faster than any classical algorithm. However,
known efficient quantum algorithms cannot find a path between the roots, as recording the
path destroys constructive interference and thus the speedup. We prove that this is inherent:
any quantum algorithm needs exponentially many queries to find a path between the roots of
an independently matched welded tree graph. This provides an example of a problem that
a quantum computer can solve exponentially faster than any classical algorithm by exploring
exponentially many paths in superposition, but where it is provably intractable to find any
such path. Theproofusescompressedpermutationoraclestorecordtheprogressofaquantum
algorithm as it queries the graph. We show that the compressed database remains path-free up
to a small error. By controlling such errors and bounding the progress of the algorithm with
each compressed oracle query, we show that Ω(2^{n/12}) queries are required to find a path in a
height-n tree with constant success probability

I am a graduate student at the University of Maryland, specifically the Joint Center for Quantum Information and Computer Science. I am fortunate to be advised by Andrew Childs.

My research addresses both the potential and limitations of quantum computers, primarily through the lens of theoretical computer science. I am especially interested in quantum query complexity, and its connections to algorithms and cryptography. A recent focus has been the quantum random oracle model, and more broadly the foundations of post-quantum cryptography.

Linnea Grans-Samuelsson, Université de Sherbrooke

Distributed architectures have been proposed as a pathway to large-scale quantum computers. Combined with the need for fault-tolerance, such architectures require distributed quantum error correction and distributed logical gates. We consider a setting where QPUs are connected using shared Bell pairs that are significantly noisier than on-chip operations. We extend the work in arXiv:2506.17181 on fault tolerance by construction to this setting, deriving different strategies for handling the additional noise. We recover conventional entanglement distillation, but also find more dynamical protocols that allow for space-time trade-offs. Through fault equivalence we show that with integrated decoding, significantly fewer Bell pairs are needed compared to entanglement distillation with separate decoding, and also that fewer Bell pairs are needed when compared to postselected entanglement distillation (repeat-until-success). As a main focus of the work, we derive efficient circuits for an important primitive in distributed fault tolerance: distributed stabilizer measurement. This primitive appears both in the context of distributed quantum memories and in the context of lattice surgery between logical qubits hosted on separate QPUs, and we showcase its usage in both settings. (We also note that the general strategies are applicable beyond stabilizer measurements, and would also apply to e.g. transversal CNOT gates.) The circuits can be adapted to resource constraints, e.g. on the Bell pair generation rate or the space available for on-chip auxiliary qubits. Noting that full local fault-tolerance is not always needed to preserve the correct scaling of logical error rates, we further optimize the circuits depending on the surrounding context. We consider in particular the surface code and the color code, both as distributed memories and in the case of lattice surgery across separate QPUs. Here, robustness to certain hook and readout errors reduces the number of Bell pairs required even further, compared to the context-free setting. We numerically benchmark the resulting implementations under circuit level noise with additional interconnect noise.

I am an incoming tenure-track assistant professor at the physics department of Université de Sherbrooke. The main focus of my research is on quantum error correction and fault-tolerant quantum computing. I am also interested in the connections between quantum error correction and statistical mechanics. Prior to obtaining my position at Sherbrooke I was (and still am, as of writing) a research fellow in theoretical physics at University of Oxford, before which I was a postdoctoral researcher at Microsoft Quantum (Station Q). I did my PhD at Institut de Physique Théorique, CEA Saclay, under the supervision of Hubert Saleur and Jesper Jacobsen, focusing on critical integrable lattice models and non-unitary conformal field theories. I am originally from Sweden.

Miryam Mi-Ying Huang, Carnegie Mellon University

Program obfuscation studies whether a program can be converted into a protected version that retains its functionality while hiding how it is implemented. In the quantum setting, obtaining such a notion for general quantum circuits has remained a longstanding challenge, and existing results have largely focused on restricted families of quantum programs.

In this talk, I will describe new constructions that extend quantum program obfuscation beyond these restricted settings to general quantum computation. I will first present an obfuscation scheme for unitary quantum programs that supports quantum inputs and outputs, thereby extending existing results for pseudo-deterministic computation. Using this construction together with a new notion of subspace-preserving pseudorandom unitaries, we further obtain a quantum ideal obfuscation scheme for arbitrary quantum circuits implementing general completely positive trace-preserving (CPTP) maps. The security of these constructions follows from the existence of post-quantum one-way functions in the classical oracle model.

Miryam Mi-Ying Huang is a postdoc research fellow at Carnegie Mellon University. Previously, she graduated from University of Southern California advised by Dr. Jiapeng Zhang. Her research primarily focuses on cryptography and complexity theory.

Bobak Kiani, Bowdoin College

Many questions in quantum information center around the expressive power of a set of quantum states: Given a Hamiltonian H, how closely can states in that class approach the ground energy of H? I will describe general methods for controlling the energy achievable by restricted classes of quantum states, thereby proving lower bounds on the resources required for state preparation. I will apply these methods in several settings including (1) no-go results for stabilizer states, (2) no-go results for states in the first level of the magic hierarchy, (3) depth lower bounds for SYK with arbitrary ancillas, and (4) tensor network bond-dimension lower bounds for spin models. This is based on joint work with Omar Al-Ghattas and David Gamarnik.

I am an assistant professor at Bowdoin College in Maine. I completed my PhD in electrical engineering and computer science at MIT. My research interests span quantum algorithms, Hamiltonian complexity, and theory of machine learning. I'm always open to meeting new people, so please come find me if you're interested in chatting with me!

Laura Lewis, UC Berkeley

We design an algorithm for learning the coefficients of an n-qubit constant-local Lindbladian to epsilon error with O(g d^2 log(n)/epsilon^2) total evolution time, where g is the single-site energy and d is the (approximate) degree of the interaction graph. Though Lindbladians present new challenges not present in the special case of Hamiltonians, our algorithm achieves the suite of desiderata attained by state-of-the-art Hamiltonian learning algorithms: (1) it uses non-adaptive, ancilla-free randomized Pauli measurement circuits with a time resolution of only Theta(1/g); (2) it works without knowledge of the structure of the unknown Lindbladian; (3) it depends on a smooth form of degree, thereby supporting the learning of quasi-local and power-law Lindbladians.

Our algorithm is a simple iterative method, where the objective function consists of Fourier coefficients of the Lindbladian restricted to few-site regions. Its analysis identifies the difficulty unique to open systems, which we call “confusing” terms. For settings where the “confusion” is limited, the performance of the algorithm improves. We demonstrate this for the case of structure learning of Hamiltonians from access to real-time evolution, where we obtain a new algorithm that is significantly simpler than previous work. In addition, using the same iterative method, we design the first efficient algorithm for structure learning Hamiltonians from high-temperature Gibbs states.

Laura is a second-year PhD student at UC Berkeley, where she is advised by Umesh Vazirani and John Wright. Previously, she was a master’s student at the University of Cambridge and the University of Edinburgh, funded by a Marshall Scholarship. Before that, she completed her bachelor’s degree at Caltech, advised by Thomas Vidick and John Preskill. Laura’s main research interests lie in quantum learning theory and quantum algorithms. 

Chaithanya Rayudu, University of Cambridge

Spectral gaps of local Hamiltonians are fundamental to dictating their physical and computational properties. A uniform lower bound on the spectral gap implies decay of correlations, stability of phases, and efficient algorithms for ground state preparation. Yet obtaining rigorous lower bounds on the spectral gap is notoriously hard, exemplified by open problems like the Haldane conjecture. In this talk, I will present a new method to obtain such bounds from extensions of the Lee-Yang theorem to quantum spin systems.

These extensions, due to Asano and Suzuki-Fisher, state that for a broad class of spin Hamiltonians on any graph, the partition function’s zeros in the complex magnetic field plane lie only on the imaginary axis. For these Hamiltonians, we prove that under a uniform Z-field of any strength h, the ground state has a spectral gap of at least h/2, independent of the system size and of the coupling strengths. The proof uses the zero-freeness of the partition function to show exponential decay of the imaginary-time correlations for any product of Z-operators. Our result gives a polynomial time quantum algorithm for computing the ground state energy of any Lee-Yang Hamiltonian.

Based on joint work (arXiv:2607.10765) with Jun Takahashi.

Chaithanya Rayudu is joining the University of Cambridge as a Postdoctoral Research Associate in fall 2026. He completed his PhD in Physics at the University of New Mexico, advised by Ojas Parekh, and holds a dual degree in Electrical Engineering from IIT Madras. His research lies at the intersection of Hamiltonian complexity, quantum algorithms, and classical optimization. His recent works are on understanding which quantum many-body systems have computationally hard ground states and thermal states, and which admit efficient algorithms.

Joseph Slote, University of Washington

State certification seeks to determine whether an approximate, lab-prepared quantum state is close to a target ideal state. There is particular interest in approaches to certification that require only simple, few-qubit measurements, and recent work has culminated in protocols certifying n-qubit quantum states with only single-qubit measurements (Huang, Preskill, Soleimanifar '24; Gupta, He, O'Donnell '25). Unfortunately, until recently these tests have been inherently non-robust: they can only positively certify lab states that are vanishingly close to the ideal state as n grows, making them difficult to apply in realistic scenarios. In this talk we describe new protocols achieving constant robustness, for example distinguishing lab states with at least 90% fidelity from those with at most 80% fidelity, independent of system size. Our protocols are based on a novel uncertainty principle for the total influence of Boolean functions and weighted generalizations thereof.

Based on joint work with Andrea Coladangelo and Jerry Li.

Joseph Slote is a postdoctoral scholar in the quantum group at the University of Washington. His research explores basic questions in quantum complexity theory with applications to learning, testing, and the certification of untrusted devices. This work enjoys a rich interplay with discrete harmonic analysis and approximation theory and has led to several new results in these fields. He earned his PhD at Caltech in 2025 under the advisement of Chris Umans and holds a master's in math and CS from the University of Oxford. Slote is a grateful recipient of NSF support through the MSPRF program.

Francisca Vasconcelos, UC Berkeley

Constant-depth quantum circuits offer a clean playground for studying the origins of quantum advantage. They are simple enough that one can hope to characterize their power, yet already exhibit phenomena with no clear classical analogue. In this talk, I will explore this question through the lens of QAC^0—the class of polynomial-size, constant-depth quantum circuits composed of arbitrary single-qubit gates and unbounded CZ/Toffoli gates. I will center the discussion around a long-standing open problem in quantum circuit complexity: is Parity computable in QAC^0? Despite its apparent simplicity, this question captures a fundamental gap in our understanding of shallow quantum computation. Resolving it would clarify how QAC^0 compares with classical AC^0 and, more broadly, how quantum circuits can use global operations such as fan-out to parallelize computation in fundamentally non-classical ways. I will survey recent upper and lower bounds on this problem and highlight how the resulting techniques have led to surprising applications, including efficient learning of QAC^0 as well as constant-depth pseudorandomness and Dicke state preparation.

Francisca Vasconcelos is a fifth-year PhD student in Computer Science at UC Berkeley, where she is advised by Michael I. Jordan and Umesh Vazirani. Her research broadly focuses on quantum algorithms and quantum complexity. Prior to Berkeley, she received her BS in Electrical Engineering, Computer Science, and Physics from MIT in 2020. She then studied at the University of Oxford as a Rhodes Scholar, completing masters degrees in Statistical Science and Philosophy of Physics. Her research experience also includes internships at Rigetti Computing, Microsoft Research Quantum, and Amazon Quantum. Francisca is a recipient of the NSF Graduate Research Fellowship and the Paul & Daisy Soros Fellowship for New Americans. She is also the Founding Academic Director of Qubit x Qubit, a global quantum computing education initiative.

Alexander Zlokapa, MIT

At large constant temperature, we show that local quantum systems with bounded interaction strength are in a separable, stabilizer thermal state within which polynomial-time classical algorithms can sample and estimate local observables. We find that the aforementioned notions of classicality break down at different temperature scales. Notably, it remains easy to classically estimate local thermal expectations at asymptotically colder temperatures than the sudden deaths of entanglement and magic. Indeed, these classical algorithms remain efficient even in the presence of large quantum circuit complexity, which we prove for the SYK model. Finally, using the non-rigorous replica trick, we conjecture that disordered all-to-all quantum systems generically experience a sharp temperature transition from classically easy to quantumly hard. Our techniques are based on constructing and sampling new cluster expansions (including for mean-field systems), zero-freeness analysis of restricted partition functions, and quantum optimal transport.

Alexander Zlokapa is currently a PhD student at MIT co-advised by Isaac Chuang and Aram Harrow, and his research interests include average-case quantum complexity, spin glass theory and quantum algorithms.