Searcharxiv⌕ Search

arXiv subjects

Yilun Xu

Publications and source records attributed to Yilun Xu.

At least 55 records · Page 3Linked to original sources
arXiv2024

Distributed Architecture for FPGA-based Superconducting Qubit Control

Quantum circuits utilizing real time feedback techniques (such as active reset and mid-circuit measurement) are a powerful tool for NISQ-era quantum computing. Such techniques are crucial for implementing error correction protocols, and can reduce the resource requirements of certain quantum algorithms. Realizing these capabilities requires flexible, low-latency classical control. We have developed a custom FPGA-based processor architecture for QubiC, an open source platform for superconducting qubit control. Our architecture is distributed in nature, and consists of a bank of lightweight cores, each configured to control a small (1-3) number of signal generator channels. Each core is capable of executing parameterized control and readout pulses, as well as performing arbitrary control flow based on mid-circuit measurement results. We have also developed a modular compiler stack and domain-specific intermediate representation for programming the processor. Our representation allows users to specify circuits using both gate and pulse-level abstractions, and includes high-level control flow constructs (e.g. if-else blocks and loops). The compiler stack is designed to integrate with quantum software tools and programming languages, such as TrueQ, pyGSTi, and OpenQASM3. In this work, we will detail the design of both the processor and compiler stack, and demonstrate its capabilities with a quantum state teleportation experiment using transmon qubits at the LBNL Advanced Quantum Testbed.

quant-ph↗
arXiv2024

QuantumLeak: Stealing Quantum Neural Networks from Cloud-based NISQ Machines

Variational quantum circuits (VQCs) have become a powerful tool for implementing Quantum Neural Networks (QNNs), addressing a wide range of complex problems. Well-trained VQCs serve as valuable intellectual assets hosted on cloud-based Noisy Intermediate Scale Quantum (NISQ) computers, making them susceptible to malicious VQC stealing attacks. However, traditional model extraction techniques designed for classical machine learning models encounter challenges when applied to NISQ computers due to significant noise in current devices. In this paper, we introduce QuantumLeak, an effective and accurate QNN model extraction technique from cloud-based NISQ machines. Compared to existing classical model stealing techniques, QuantumLeak improves local VQC accuracy by 4.99\%$\sim$7.35\% across diverse datasets and VQC architectures.

quant-ph↗
arXiv2024

TITAN: A Distributed Large-Scale Trapped-Ion NISQ Computer

Trapped-Ion (TI) technology offers potential breakthroughs for Noisy Intermediate Scale Quantum (NISQ) computing. TI qubits offer extended coherence times and high gate fidelity, making them appealing for large-scale NISQ computers. Constructing such computers demands a distributed architecture connecting Quantum Charge Coupled Devices (QCCDs) via quantum matter-links and photonic switches. However, current distributed TI NISQ computers face hardware and system challenges. Entangling qubits across a photonic switch introduces significant latency, while existing compilers generate suboptimal mappings due to their unawareness of the interconnection topology. In this paper, we introduce TITAN, a large-scale distributed TI NISQ computer, which employs an innovative photonic interconnection design to reduce entanglement latency and an advanced partitioning and mapping algorithm to optimize matter-link communications. Our evaluations show that TITAN greatly enhances quantum application performance by 56.6% and fidelity by 19.7% compared to existing systems.

quant-ph↗
arXiv2023

A Survey on Generative Diffusion Model

Deep generative models have unlocked another profound realm of human creativity. By capturing and generalizing patterns within data, we have entered the epoch of all-encompassing Artificial Intelligence for General Creativity (AIGC). Notably, diffusion models, recognized as one of the paramount generative models, materialize human ideation into tangible instances across diverse domains, encompassing imagery, text, speech, biology, and healthcare. To provide advanced and comprehensive insights into diffusion, this survey comprehensively elucidates its developmental trajectory and future directions from three distinct angles: the fundamental formulation of diffusion, algorithmic enhancements, and the manifold applications of diffusion. Each layer is meticulously explored to offer a profound comprehension of its evolution. Structured and summarized approaches are presented in https://github.com/chq1155/A-Survey-on-Generative-Diffusion-Model.

cs.AI↗
arXiv2023

New Facts and Data about Professors and their Research

We introduce a new survey of professors at roughly 150 of the most research-intensive institutions of higher education in the US. We document seven new features of how research-active professors are compensated, how they spend their time, and how they perceive their research pursuits: (1) there is more inequality in earnings within fields than there is across fields; (2) institutions, ranks, tasks, and sources of earnings can account for roughly half of the total variation in earnings; (3) there is significant variation across fields in the correlations between earnings and different kinds of research output, but these account for a small amount of earnings variation; (4) measuring professors' productivity in terms of output-per-year versus output-per-research-hour can yield substantial differences; (5) professors' beliefs about the riskiness of their research are best predicted by their fundraising intensity, their risk-aversion in their personal lives, and the degree to which their research involves generating new hypotheses; (6) older and younger professors have very different research outputs and time allocations, but their intended audiences are quite similar; (7) personal risk-taking is highly predictive of professors' orientation towards applied, commercially-relevant research.

econ.GN↗
arXiv2023

Particle Guidance: non-I.I.D. Diverse Sampling with Diffusion Models

In light of the widespread success of generative models, a significant amount of research has gone into speeding up their sampling time. However, generative models are often sampled multiple times to obtain a diverse set incurring a cost that is orthogonal to sampling time. We tackle the question of how to improve diversity and sample efficiency by moving beyond the common assumption of independent samples. We propose particle guidance, an extension of diffusion-based generative sampling where a joint-particle time-evolving potential enforces diversity. We analyze theoretically the joint distribution that particle guidance generates, how to learn a potential that achieves optimal diversity, and the connections with methods in other disciplines. Empirically, we test the framework both in the setting of conditional image generation, where we are able to increase diversity without affecting quality, and molecular conformer generation, where we reduce the state-of-the-art median error by 13% on average.

cs.LG↗
arXiv2023

Restart Sampling for Improving Generative Processes

Generative processes that involve solving differential equations, such as diffusion models, frequently necessitate balancing speed and quality. ODE-based samplers are fast but plateau in performance while SDE-based samplers deliver higher sample quality at the cost of increased sampling time. We attribute this difference to sampling errors: ODE-samplers involve smaller discretization errors while stochasticity in SDE contracts accumulated errors. Based on these findings, we propose a novel sampling algorithm called Restart in order to better balance discretization errors and contraction. The sampling method alternates between adding substantial noise in additional forward steps and strictly following a backward ODE. Empirically, Restart sampler surpasses previous SDE and ODE samplers in both speed and accuracy. Restart not only outperforms the previous best SDE results, but also accelerates the sampling speed by 10-fold / 2-fold on CIFAR-10 / ImageNet $64 \times 64$. In addition, it attains significantly better sample quality than ODE samplers within comparable sampling times. Moreover, Restart better balances text-image alignment/visual quality versus diversity than previous samplers in the large-scale text-to-image Stable Diffusion model pre-trained on LAION $512 \times 512$. Code is available at https://github.com/Newbeeer/diffusion_restart_sampling

cs.LG↗
arXiv2023

Fast quantum state transfer and entanglement preparation in strongly coupled bosonic systems

Continuous U(1) gauge symmetry, which guarantees the conservation of the total excitations in linear bosonic systems, will be broken when it comes to the strong-coupling regime where the rotation wave approximation (RWA) fails. Here we develop analytic solutions for multi-mode bosonic systems with XX-type couplings beyond RWA, and proposed a novel scheme to implement high-fidelity quantum state transfer (QST) and entanglement preparation (EP) with high speed. The scheme can be realized with designated coupling strength and pulse duration with which the excitation number keeps unchanged regardless of the breakdown of the global U(1) symmetry. In the QST tasks, we consider several typical quantum states and demonstrate that this method is robust against thermal noise and imperfections of experimental sequence. In the EP tasks, the scheme is successfully implemented for the preparation of Bell states and W-type states, within a shortest preparation time.

quant-ph↗
arXiv2023

QubiC 2.0: An Extensible Open-Source Qubit Control System Capable of Mid-Circuit Measurement and Feed-Forward

Researchers manipulate and measure quantum processing units via the classical electronics control system. We developed an open-source FPGA-based quantum bit control system called QubiC for superconducting qubits. After a few years of qubit calibration and testing experience on QubiC 1.0, we recognized the need for mid-circuit measurements and feed-forward capabilities to implement advanced quantum algorithms effectively. Moreover, following the development of RFSoC technology, we upgraded the system to QubiC 2.0 on an Xilinx ZCU216 evaluation board and developed all these enriched features. The system uses portable FPGA gateware with a simplified processor to handle commands on-the-fly. For design simplicity and straightforward scaling, we adopted a multi-core distributed architecture, assigning one processor core per qubit. The actual pulses combine the unique pulse envelope and carrier information specified in a command. Each pulse envelope is pre-stored on FPGA's block RAMs, ensuring the speed and reusability during the whole circuit. The pulse parameters including amplitude, phase, and frequency can be updated from pulse to pulse. The software stack is developed in Python, running on both the FPGA's ARM core and host computer via XML-RPC. The quantum circuit can be described in a high-level language, which supports programming at both pulse-level and native-gate level, and includes high-level control flow constructs. The QubiC software stack compiles these quantum programs into binary commands that can be loaded into the FPGA. With Qubic 2.0, we successfully achieved multi-FPGA synchronization in bench tests and demonstrated simplified feed-forward experiments on conditional circuits. The enhanced QubiC system represents a significant step forward in quantum computing, providing researchers with powerful tools to explore and implement advanced quantum algorithms and applications.

quant-ph↗
arXiv2023

Slip intermittency and dwell fatigue in titanium alloys: a discrete dislocation plasticity analysis

Slip intermittency and stress oscillations in titanium alloy Ti-7Al-O that were observed using in-situ far-field high energy X-ray diffraction microscopy (ff-HEDM) are investigated using a discrete dislocation plasticity (DDP) model. The mechanistic foundation of slip intermittency and stress oscillations are shown to be dislocation escape from obstacles during stress holds, governed by a thermal activation constitutive law. The stress drop events due to -basal slip are larger in magnitude than those along -prism, which is a consequence of their differing rate sensitivities, previously found from micropillar testing. It is suggested that interstitial oxygen suppresses stress oscillations by inhibiting the thermal activation process. Understanding of these mechanisms is of benefit to the design and safety assessment of jet engine titanium alloys subjected to dwell fatigue.

cond-mat.mtrl-sci↗
arXiv2023

GenPhys: From Physical Processes to Generative Models

Since diffusion models (DM) and the more recent Poisson flow generative models (PFGM) are inspired by physical processes, it is reasonable to ask: Can physical processes offer additional new generative models? We show that the answer is yes. We introduce a general family, Generative Models from Physical Processes (GenPhys), where we translate partial differential equations (PDEs) describing physical processes to generative models. We show that generative models can be constructed from s-generative PDEs (s for smooth). GenPhys subsume the two existing generative models (DM and PFGM) and even give rise to new families of generative models, e.g., "Yukawa Generative Models" inspired from weak interactions. On the other hand, some physical processes by default do not belong to the GenPhys family, e.g., the wave equation and the Schrödinger equation, but could be made into the GenPhys family with some modifications. Our goal with GenPhys is to explore and expand the design space of generative models.

cs.LG↗
arXiv2023

Stable Target Field for Reduced Variance Score Estimation in Diffusion Models

Diffusion models generate samples by reversing a fixed forward diffusion process. Despite already providing impressive empirical results, these diffusion models algorithms can be further improved by reducing the variance of the training targets in their denoising score-matching objective. We argue that the source of such variance lies in the handling of intermediate noise-variance scales, where multiple modes in the data affect the direction of reverse paths. We propose to remedy the problem by incorporating a reference batch which we use to calculate weighted conditional scores as more stable training targets. We show that the procedure indeed helps in the challenging intermediate regime by reducing (the trace of) the covariance of training targets. The new stable targets can be seen as trading bias for reduced variance, where the bias vanishes with increasing reference batch size. Empirically, we show that the new objective improves the image quality, stability, and training speed of various popular diffusion models across datasets with both general ODE and SDE solvers. When used in combination with EDM, our method yields a current SOTA FID of 1.90 with 35 network evaluations on the unconditional CIFAR-10 generation task. The code is available at https://github.com/Newbeeer/stf

cs.LG↗
arXiv2023

PFGM++: Unlocking the Potential of Physics-Inspired Generative Models

We introduce a new family of physics-inspired generative models termed PFGM++ that unifies diffusion models and Poisson Flow Generative Models (PFGM). These models realize generative trajectories for $N$ dimensional data by embedding paths in $N{+}D$ dimensional space while still controlling the progression with a simple scalar norm of the $D$ additional variables. The new models reduce to PFGM when $D{=}1$ and to diffusion models when $D{\to}\infty$. The flexibility of choosing $D$ allows us to trade off robustness against rigidity as increasing $D$ results in more concentrated coupling between the data and the additional variable norms. We dispense with the biased large batch field targets used in PFGM and instead provide an unbiased perturbation-based objective similar to diffusion models. To explore different choices of $D$, we provide a direct alignment method for transferring well-tuned hyperparameters from diffusion models ($D{\to} \infty$) to any finite $D$ values. Our experiments show that models with finite $D$ can be superior to previous state-of-the-art diffusion models on CIFAR-10/FFHQ $64{\times}64$ datasets, with FID scores of $1.91/2.43$ when $D{=}2048/128$. In class-conditional setting, $D{=}2048$ yields current state-of-the-art FID of $1.74$ on CIFAR-10. In addition, we demonstrate that models with smaller $D$ exhibit improved robustness against modeling errors. Code is available at https://github.com/Newbeeer/pfgmpp

cs.LG↗
arXiv2023

Novel Intensity Mapping Functions: Weighted Histogram Averaging

It is challenging to align the brightness distribution of the images with different exposures due to possible color distortion and loss of details in the brightest and darkest regions of input images. In this paper, a novel intensity mapping algorithm is first proposed by introducing a new concept of weighted histogram averaging (WHA). The proposed WHA algorithm leverages the correspondence between the histogram bins of two images which are built up by using the non-decreasing property of the intensity mapping functions (IMFs). Extensive experiments indicate that the proposed WHA algorithm significantly surpasses the related state-of-the-art intensity mapping methods.

cs.CV↗
arXiv2022

Poisson Flow Generative Models

We propose a new "Poisson flow" generative model (PFGM) that maps a uniform distribution on a high-dimensional hemisphere into any data distribution. We interpret the data points as electrical charges on the $z=0$ hyperplane in a space augmented with an additional dimension $z$, generating a high-dimensional electric field (the gradient of the solution to Poisson equation). We prove that if these charges flow upward along electric field lines, their initial distribution in the $z=0$ plane transforms into a distribution on the hemisphere of radius $r$ that becomes uniform in the $r \to\infty$ limit. To learn the bijective transformation, we estimate the normalized field in the augmented space. For sampling, we devise a backward ODE that is anchored by the physically meaningful additional dimension: the samples hit the unaugmented data manifold when the $z$ reaches zero. Experimentally, PFGM achieves current state-of-the-art performance among the normalizing flow models on CIFAR-10, with an Inception score of $9.68$ and a FID score of $2.35$. It also performs on par with the state-of-the-art SDE approaches while offering $10\times $ to $20 \times$ acceleration on image generation tasks. Additionally, PFGM appears more tolerant of estimation errors on a weaker network architecture and robust to the step size in the Euler method. The code is available at https://github.com/Newbeeer/poisson_flow .

cs.LG↗
arXiv2022

Multi-scale plasticity homogenization of Sn-3Ag-0.5Cu: from β-Sn micropillars to polycrystals with intermetallics

The mechanical properties of $β$-Sn single crystals have been systematically investigated using a combined methodology of micropillar tests and rate-dependent crystal plasticity modelling. The slip strength and rate sensitivity of several key slip systems within $β$-Sn single crystals have been determined. Consistency between the numerically predicted and experimentally observed slip traces has been shown for pillars oriented to activate single and double slip. Subsequently, the temperature-dependent, intermetallic-size-governing behaviour of a polycrystal $β$-Sn-rich alloy SAC305 (96.5Sn-3Ag-0.5Cu wt%) is predicted through a multi-scale homogenization approach, and the predicted temperature- and rate-sensitivity reproduce independent experimental results. The integrated experimental and numerical approaches provide mechanistic understanding and fundamental material properties of microstructure-sensitive behaviour of electronic solders subject to thermomechanical loading, including thermal fatigue.

cond-mat.mtrl-sci↗
arXiv2022

The influence of alloying on slip intermittency and the implications for dwell fatigue in titanium

Dwell fatigue, the reduction in fatigue life experienced by titanium alloys due to holds at stresses as low as 60% of yield, has been implicated in several uncontained jet engine failures. Dislocation slip has long been observed to be an intermittent, scale-bridging phenomenon, similar to that seen in earthquakes but at the nanoscale, leading to the speculation that large stress bursts might promote the initial opening of a crack. Here we observe such stress bursts at the scale of individual grains in situ, using high energy X-ray diffraction microscopy in Ti-7Al-O alloys. This shows that the detrimental effect of precipitation of ordered Ti_3Al is to increase the magnitude of rare pri and bas slip bursts associated with slip localisation. In contrast, the addition of trace O interstitials is beneficial, reducing the magnitude of bas slip bursts and increasing the homogeneity between basal and prismatic slip. This is further evidence that the formation of long paths for easy basal plane slip localisation should be avoided when engineering titanium alloys against dwell fatigue.

cond-mat.mtrl-sci↗
arXiv2022

A multi-scale approach to microstructure-sensitive thermal fatigue in solder joints

This paper presents a multi-scale modelling approach to investigate the underpinning mechanisms of microstructure-sensitive damage of single crystal Sn-3Ag-0.5Cu (wt%, SAC305) solder joints of a Ball Grid Array (BGA) board assembly subject to thermal cycling. The multi-scale scheme couples board-scale modelling at the continuum macro-scale and individual solder modelling at the crystal micro-scale. Systematic studies of tin crystal orientation and its role in fatigue damage have been compared to experimental observations. Crystallographic orientation is examined with respect to damage development, providing evidence-based optimal solder microstructural design for in-service thermomechanical fatigue.

cond-mat.mtrl-sci↗