Results 851 - 860 of 24440
In this talk, I will introduce a holographic framework for analyzing the steady states of repeated quantum channels. Using channel–state duality, we show that the steady state of a $d$-dimensional quantum channel is holographically mapped to the boundary reduced density matrix of a $(d+1)$-dimensional wavefunction generated by a sequential unitary circuit. From this perspective, strong-to-weak spontaneous symmetry breaking (SWSSB) in the steady state arises from the anyon condensation on the boundary of a topological order in one higher dimension. The conditional mutual information (CMI) associated with SWSSB is then inherited from the bulk topological entanglement entropy. We make this duality explicit using isometric tensor network states (isoTNS) by identifying the channel’s time evolution with the transfer matrix of a higher-dimensional isoTNS. Built on isoTNS, we further construct continuously tunable quantum channels that exhibit distinct mixed-state phases and transitions in the steady states.
Quantum metrology involves the application of quantum resources to enhance measurements. Key challenges include preparing metrologically useful states, maintaining coherence during sensing, and efficiently extracting information from quantum systems. I will overview the basics of quantum metrology, recent advances, and discuss emerging approaches that extend the traditional metrology paradigm through time reversal, measurement, and feedback. I will present recent work demonstrating new sensing modalities enabled by programmable quantum circuits. Protocols inspired by simulations of time-travel enable optimal phase estimation without needing to know exactly the unitary causing a rotation. Platform-specific unitary inversion realizes new advantages in terms of quantum resource demands. Measurement-conditioned dynamics can efficiently generate entangled states through feedforward control. As quantum systems scale, interactive approaches combining measurement, feedback, and adaptive control may offer practical advantages for preparing and maintaining metrologically relevant quantum states. I will discuss prospects and open questions for this emerging paradigm.
Matei Zaharia, associate professor of electrical engineering and computer sciences (EECS) at UC Berkeley, has been awarded the ACM Prize in Computing for his visionary development of distributed data systems and computing infrastructure. In the prize announcement, the Association for Computing Machinery (ACM) noted Zaharia’s development of open-source systems helped enable large-scale machine learning (ML), analytics and AI at a global scale.
The field of superconducting circuits has emerged as a rich platform for both quantum computation and quantum simulation. Due to the native strong qubit-photon interactions and highly controlled dissipation, these systems can be used to study dynamical phase transitions, many-body phenomena, and spin models in driven-dissipative systems. Temporal control can be used to implement synthetic dimensions and Floquet lattice systems. I will present data from two experiments which realize autonomous stabilization of time-dependent Floquet states and the first direct observation of a topological photon pump.
I will discuss a connection between mixed-state phase transitions in open quantum dynamics, their associated learning problems, and their relation to modern generative models, in particular diffusion models. Recent progress suggests that mixed-state phases can be characterized through local reversibility. On the classical side, a similar phenomenon appears in diffusion models: away from phase transitions, local neural networks are sufficient and lead to more efficient learning and training. On the quantum side, this perspective leads to a learning problem where phase transitions can be avoided along certain dynamical paths. In this case, we construct efficient learning algorithms for both sample and computational complexity.
These parallels suggest that learning processes can be understood in terms of phase structure, with phase transitions marking the breakdown of Markovianity. Finally, I will outline several open questions along this direction, including mechanically interpretable diffusion models inspired by locality and Markovianity, and a possible way of defining quantum advantage for generating complex probability distributions based on mixed-state phase transitions and their underlying physical mechanisms.