Searcharxiv⌕ Search

arXiv subjects

Yu Chen

Publications and source records attributed to Yu Chen.

At least 307 records · Page 17Linked to original sources
arXiv2024

Sample-based Dynamic Hierarchical Transformer with Layer and Head Flexibility via Contextual Bandit

Transformer requires a fixed number of layers and heads which makes them inflexible to the complexity of individual samples and expensive in training and inference. To address this, we propose a sample-based Dynamic Hierarchical Transformer (DHT) model whose layers and heads can be dynamically configured with single data samples via solving contextual bandit problems. To determine the number of layers and heads, we use the Uniform Confidence Bound while we deploy combinatorial Thompson Sampling in order to select specific head combinations given their number. Different from previous work that focuses on compressing trained networks for inference only, DHT is not only advantageous for adaptively optimizing the underlying network architecture during training but also has a flexible network for efficient inference. To the best of our knowledge, this is the first comprehensive data-driven dynamic transformer without any additional auxiliary neural networks that implement the dynamic system. According to the experiment results, we achieve up to 74% computational savings for both training and inference with a minimal loss of accuracy.

cs.LG↗
arXiv2024

Optimizing quantum gates towards the scale of logical qubits

A foundational assumption of quantum error correction theory is that quantum gates can be scaled to large processors without exceeding the error-threshold for fault tolerance. Two major challenges that could become fundamental roadblocks are manufacturing high performance quantum hardware and engineering a control system that can reach its performance limits. The control challenge of scaling quantum gates from small to large processors without degrading performance often maps to non-convex, high-constraint, and time-dependent control optimization over an exponentially expanding configuration space. Here we report on a control optimization strategy that can scalably overcome the complexity of such problems. We demonstrate it by choreographing the frequency trajectories of 68 frequency-tunable superconducting qubits to execute single- and two-qubit gates while mitigating computational errors. When combined with a comprehensive model of physical errors across our processor, the strategy suppresses physical error rates by $\sim3.7\times$ compared with the case of no optimization. Furthermore, it is projected to achieve a similar performance advantage on a distance-23 surface code logical qubit with 1057 physical qubits. Our control optimization strategy solves a generic scaling challenge in a way that can be adapted to a variety of quantum operations, algorithms, and computing architectures.

quant-ph↗
arXiv2024

Investigation for $D^+ \to π^+ ν\barν$ decay process within QCDSR approach

In the paper, we investigate the charmed meson rare decay process $D^+ \to π^+ν\barν$ by using QCD sum rules approach. Firstly, the pion twist-2 and twist-3 distribution amplitude $ξ$-moments $\langleξ_{2;π}^n\rangle|_μ$ up to 10th-order and $\langle ξ_{3;π}^{(p,σ),n}\rangle|_μ$ up to fourth-order are calculated by using QCD sum rule under background field theory. After constructing the light-cone harmonic oscillator model for pion twist-2, 3 DAs, we get their behaviors by matching the calculated $ξ$-moments. Then, the $D\to π$ transition form factors are calculated by using QCD light-cone sum rules approach. The vector form factor at large recoil region is $f_+^{D\toπ}(0) = 0.627^{+0.120} _{-0.080}$. By taking the rapidly $z(q^2,t)$ converging simplified series expansion, we present the TFFs and the corresponding angular coefficients in the whole squared momentum transfer physical region. Furthermore, we display the semileptonic decay process $\bar D^0 \to π^+ e\bar ν_e$ differential decay widths and branching fraction with ${\cal B}(\bar D^0\toπ^+e\barν_e) = 0.308^{+0.155}_{-0.066} \times 10^{2}$. The $\bar D^0\toπ^+e\barν_e$ differential/total predictions for forward-backward asymmetry, $q^2$-differential flat terms and lepton polarization asymmetry are also given. After considering the non-standard neutrino interactions, the predictions for the $D^+ \to π^+ ν\barν$ branching fraction is ${\cal B}(D^+ \to π^+ {ν}{\barν}) = 1.85^{+0.93}_{-0.46}\times10^{-8}$.

hep-ph↗
arXiv2024

Reflected Schrödinger Bridge for Constrained Generative Modeling

Diffusion models have become the go-to method for large-scale generative models in real-world applications. These applications often involve data distributions confined within bounded domains, typically requiring ad-hoc thresholding techniques for boundary enforcement. Reflected diffusion models (Lou23) aim to enhance generalizability by generating the data distribution through a backward process governed by reflected Brownian motion. However, reflected diffusion models may not easily adapt to diverse domains without the derivation of proper diffeomorphic mappings and do not guarantee optimal transport properties. To overcome these limitations, we introduce the Reflected Schrodinger Bridge algorithm: an entropy-regularized optimal transport approach tailored for generating data within diverse bounded domains. We derive elegant reflected forward-backward stochastic differential equations with Neumann and Robin boundary conditions, extend divergence-based likelihood training to bounded domains, and explore natural connections to entropic optimal transport for the study of approximate linear convergence - a valuable insight for practical training. Our algorithm yields robust generative modeling in diverse domains, and its scalability is demonstrated in real-world constrained generative modeling through standard image benchmarks.

stat.ML↗
arXiv2024

Correlated sensing with a solid-state quantum multi-sensor system for atomic-scale structural analysis

Developing superior quantum sensing strategies ranging from ultra-high precision measurement to complex structural analysis is at the heart of quantum technologies. While strategies using quantum resources, such as entanglement among sensors, to enhance the sensing precision have been abundantly demonstrated, the signal correlation among quantum sensors is rarely exploited. Here we develop a novel sensing paradigm exploiting the signal correlation among multiple quantum sensors to resolve overlapping signals from multiple targets that individual sensors can't resolve and complex structural construction struggles with. With three nitrogen-vacancy centers as a quantum electrometer system, we demonstrate this multi-sensor paradigm by resolving individual defects' fluctuating electric fields from ensemble signals. We image the three-dimensional distribution of 16 dark electronic point-defects in diamond with accuracy approaching 1.7 nm via a GPS-like localization method. Furthermore, we obtain the real-time charge dynamics of individual point defects and visualize how the dynamics induce the well-known optical spectral diffusion. The multi-sensor paradigm extends the quantum sensing toolbox and offers new possibilities for structural analysis.

quant-ph↗
arXiv2023

SegRap2023: A Benchmark of Organs-at-Risk and Gross Tumor Volume Segmentation for Radiotherapy Planning of Nasopharyngeal Carcinoma

Radiation therapy is a primary and effective NasoPharyngeal Carcinoma (NPC) treatment strategy. The precise delineation of Gross Tumor Volumes (GTVs) and Organs-At-Risk (OARs) is crucial in radiation treatment, directly impacting patient prognosis. Previously, the delineation of GTVs and OARs was performed by experienced radiation oncologists. Recently, deep learning has achieved promising results in many medical image segmentation tasks. However, for NPC OARs and GTVs segmentation, few public datasets are available for model development and evaluation. To alleviate this problem, the SegRap2023 challenge was organized in conjunction with MICCAI2023 and presented a large-scale benchmark for OAR and GTV segmentation with 400 Computed Tomography (CT) scans from 200 NPC patients, each with a pair of pre-aligned non-contrast and contrast-enhanced CT scans. The challenge's goal was to segment 45 OARs and 2 GTVs from the paired CT scans. In this paper, we detail the challenge and analyze the solutions of all participants. The average Dice similarity coefficient scores for all submissions ranged from 76.68\% to 86.70\%, and 70.42\% to 73.44\% for OARs and GTVs, respectively. We conclude that the segmentation of large-size OARs is well-addressed, and more efforts are needed for GTVs and small-size or thin-structure OARs. The benchmark will remain publicly available here: https://segrap2023.grand-challenge.org

eess.IV↗
arXiv2023

Polarization Criterion in Targeted SETI Observation

In this paper, we propose a novel method for distinguishing extraterrestrial intelligence (ETI) signals from radio frequency interference (RFI) by leveraging polarization features. We exploit the sinusoidal variation of the linearly polarized components of Stokes parameters with the parallactic angle as a characteristic signature of ETI signals, while such linearly polarized components remain relatively stable for terrestrial RFI. Typically, a minimum of 4-8 hours of observation time is required to detect these sinusoidal variations. The polarization approach in the search for extraterrestrial intelligence (SETI) also enables us to study the radio stellar bursts emitted by M-type stars as ancillary science, which is relevant to assessing the habitability of exoplanets. Compared to the frequency drift method, the polarization method effectively reduces the required observation time for signal identification while improving the signal identification process.

astro-ph.IM↗
arXiv2023

Provably Efficient Iterated CVaR Reinforcement Learning with Function Approximation and Human Feedback

Risk-sensitive reinforcement learning (RL) aims to optimize policies that balance the expected reward and risk. In this paper, we present a novel risk-sensitive RL framework that employs an Iterated Conditional Value-at-Risk (CVaR) objective under both linear and general function approximations, enriched by human feedback. These new formulations provide a principled way to guarantee safety in each decision making step throughout the control process. Moreover, integrating human feedback into risk-sensitive RL framework bridges the gap between algorithmic decision-making and human participation, allowing us to also guarantee safety for human-in-the-loop systems. We propose provably sample-efficient algorithms for this Iterated CVaR RL and provide rigorous theoretical analysis. Furthermore, we establish a matching lower bound to corroborate the optimality of our algorithms in a linear context.

cs.LG↗
arXiv2023

Real-time Monitoring for the Next Core-Collapse Supernova in JUNO

The core-collapse supernova (CCSN) is considered one of the most energetic astrophysical events in the universe. The early and prompt detection of neutrinos before (pre-SN) and during the supernova (SN) burst presents a unique opportunity for multi-messenger observations of CCSN events. In this study, we describe the monitoring concept and present the sensitivity of the system to pre-SN and SN neutrinos at the Jiangmen Underground Neutrino Observatory (JUNO), a 20 kton liquid scintillator detector currently under construction in South China. The real-time monitoring system is designed to ensure both prompt alert speed and comprehensive coverage of progenitor stars. It incorporates prompt monitors on the electronic board as well as online monitors at the data acquisition stage. Assuming a false alert rate of 1 per year, this monitoring system exhibits sensitivity to pre-SN neutrinos up to a distance of approximately 1.6 (0.9) kiloparsecs and SN neutrinos up to about 370 (360) kiloparsecs for a progenitor mass of 30 solar masses, considering both normal and inverted mass ordering scenarios. The pointing ability of the CCSN is evaluated by analyzing the accumulated event anisotropy of inverse beta decay interactions from pre-SN or SN neutrinos. This, along with the early alert, can play a crucial role in facilitating follow-up multi-messenger observations of the next galactic or nearby extragalactic CCSN.

hep-ex↗
arXiv2023

A causal convolutional neural network for multi-subject motion modeling and generation

Inspired by the success of WaveNet in multi-subject speech synthesis, we propose a novel neural network based on causal convolutions for multi-subject motion modeling and generation. The network can capture the intrinsic characteristics of the motion of different subjects, such as the influence of skeleton scale variation on motion style. Moreover, after fine-tuning the network using a small motion dataset for a novel skeleton that is not included in the training dataset, it is able to synthesize high-quality motions with a personalized style for the novel skeleton. The experimental results demonstrate that our network can model the intrinsic characteristics of motions well and can be applied to various motion modeling and synthesis tasks.

cs.GR↗
arXiv2023

A novel decomposed-ensemble time series forecasting framework: capturing underlying volatility information

Time series forecasting represents a significant and challenging task across various fields. Recently, methods based on mode decomposition have dominated the forecasting of complex time series because of the advantages of capturing local characteristics and extracting intrinsic modes from data. Unfortunately, most models fail to capture the implied volatilities that contain significant information. To enhance the prediction of contemporary diverse and complex time series, we propose a novel time series forecasting paradigm that integrates decomposition with the capability to capture the underlying fluctuation information of the series. In our methodology, we implement the Variational Mode Decomposition algorithm to decompose the time series into K distinct sub-modes. Following this decomposition, we apply the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model to extract the volatility information in these sub-modes. Subsequently, both the numerical data and the volatility information for each sub-mode are harnessed to train a neural network. This network is adept at predicting the information of the sub-modes, and we aggregate the predictions of all sub-modes to generate the final output. By integrating econometric and artificial intelligence methods, and taking into account both the numerical and volatility information of the time series, our proposed framework demonstrates superior performance in time series forecasting, as evidenced by the significant decrease in MSE, RMSE, and MAPE in our comparative experimental results.

stat.ME↗
arXiv2023

Interpreting Differentiable Latent States for Healthcare Time-series Data

Machine learning enables extracting clinical insights from large temporal datasets. The applications of such machine learning models include identifying disease patterns and predicting patient outcomes. However, limited interpretability poses challenges for deploying advanced machine learning in digital healthcare. Understanding the meaning of latent states is crucial for interpreting machine learning models, assuming they capture underlying patterns. In this paper, we present a concise algorithm that allows for i) interpreting latent states using highly related input features; ii) interpreting predictions using subsets of input features via latent states; and iii) interpreting changes in latent states over time. The proposed algorithm is feasible for any model that is differentiable. We demonstrate that this approach enables the identification of a daytime behavioral pattern for predicting nocturnal behavior in a real-world healthcare dataset.

cs.LG↗
arXiv2023

SCALAR-NeRF: SCAlable LARge-scale Neural Radiance Fields for Scene Reconstruction

In this work, we introduce SCALAR-NeRF, a novel framework tailored for scalable large-scale neural scene reconstruction. We structure the neural representation as an encoder-decoder architecture, where the encoder processes 3D point coordinates to produce encoded features, and the decoder generates geometric values that include volume densities of signed distances and colors. Our approach first trains a coarse global model on the entire image dataset. Subsequently, we partition the images into smaller blocks using KMeans with each block being modeled by a dedicated local model. We enhance the overlapping regions across different blocks by scaling up the bounding boxes of each local block. Notably, the decoder from the global model is shared across distinct blocks and therefore promoting alignment in the feature space of local encoders. We propose an effective and efficient methodology to fuse the outputs from these local models to attain the final reconstruction. Employing this refined coarse-to-fine strategy, our method outperforms state-of-the-art NeRF methods and demonstrates scalability for large-scale scene reconstruction. The code will be available on our project page at https://aibluefisher.github.io/SCALAR-NeRF/

cs.CV↗
arXiv2023

Cooperation between coherent control and noises in quantum metrology

In this paper we study the cooperation between coherent control and noises in open spin systems, aiming to demonstrate that such cooperation can provide new possibilities for parametrization in quantum metrology. The cooperative scheme proposed here outperforms the standard scheme, with higher precision achieved. More specifically, we illustrate the effect of cooperative interaction between coherent control and noises in conventional single-spin systems described by Lindblad master equations, with the magnitude of a magnetic field taken as the parameter to be estimated and encoded in the noisy dynamics besides the Hamiltonian. The scenarios of both spontaneous emission and dephasing noise have been analyzed, where the associated quantum Fisher information has been given. Furthermore, it has been demonstrated that in the realm where quantum metrology is mostly applied in practice, the precision limit under the cooperative scheme in the presence of noises can surpass the ultimate precision limit under the unitary dynamics. On the other hand, multiple-spin systems have also been considered. We show that the coupling between different spins can help realize non-local parametrization under the cooperative scheme, with the ground state made entangled. This thus leads to the improvement of precision limit, which has been proved and visualized in our paper.

quant-ph↗
arXiv2023

Scaling Law of Large Sequential Recommendation Models

Scaling of neural networks has recently shown great potential to improve the model capacity in various fields. Specifically, model performance has a power-law relationship with model size or data size, which provides important guidance for the development of large-scale models. However, there is still limited understanding on the scaling effect of user behavior models in recommender systems, where the unique data characteristics (e.g. data scarcity and sparsity) pose new challenges to explore the scaling effect in recommendation tasks. In this work, we focus on investigating the scaling laws in large sequential recommendation models. Specially, we consider a pure ID-based task formulation, where the interaction history of a user is formatted as a chronological sequence of item IDs. We don't incorporate any side information (e.g. item text), because we would like to explore how scaling law holds from the perspective of user behavior. With specially improved strategies, we scale up the model size to 0.8B parameters, making it feasible to explore the scaling effect in a diverse range of model sizes. As the major findings, we empirically show that scaling law still holds for these trained models, even in data-constrained scenarios. We then fit the curve for scaling law, and successfully predict the test loss of the two largest tested model scales. Furthermore, we examine the performance advantage of scaling effect on five challenging recommendation tasks, considering the unique issues (e.g. cold start, robustness, long-term preference) in recommender systems. We find that scaling up the model size can greatly boost the performance on these challenging tasks, which again verifies the benefits of large recommendation models.

cs.IR↗
arXiv2023

A Closer Look at Small-Scale Magnetic Flux Ropes in the Solar Wind at 1 AU: Results from Improved Automated Detection

Small-scale interplanetary magnetic flux ropes (SMFRs) are similar to ICMEs in magnetic structure, but are smaller and do not exhibit ICME plasma signatures. We present a computationally efficient and GPU-powered version of the single-spacecraft automated SMFR detection algorithm based on the Grad-Shafranov (GS) technique. Our algorithm is capable of processing higher resolution data, eliminates selection bias caused by a fixed $\avg{B}$ threshold, has improved detection criteria demonstrated to have better results on an MHD simulation, and recovers full 2.5D cross sections using GS reconstruction. We used it to detect 512,152 SMFRs from 27 years (1996 to 2022) of 3-second cadence \emph{Wind} measurements. Our novel findings are: (1) the radial density of SMFRs at 1 au (${\sim}1$ per $\si{10^6\kilo\meter}$) and filling factor (${\sim}$35\%) are independent of solar activity, distance to the heliospheric current sheet (HCS), and solar wind plasma type, although the minority of SMFRs with diameters greater than ${\sim}$0.01 au have a strong solar activity dependence; (2) SMFR diameters follow a log-normal distribution that peaks below the resolved range ($\gtrsim 10^4$ km), although the filling factor is dominated by SMFRs between $10^5$ to $10^6$ km; (3) most SMFRs at 1 au have strong field-aligned flows like those from PSP measurements; (4) in terms of diameter $d$, SMFR poloidal flux $\propto d^{1.2}$, axial flux $\propto d^{2.0}$, average twist number $\propto d^{-0.8}$, current density $\propto d^{-0.8}$, and helicity $\propto d^{3.2}$. Implications for the origin of SMFRs and switchbacks are briefly discussed. The new algorithm and SMFR dataset are made freely available.

physics.space-ph↗
arXiv2023

Measurement-Induced State Transitions in a Superconducting Qubit: Within the Rotating Wave Approximation

Superconducting qubits typically use a dispersive readout scheme, where a resonator is coupled to a qubit such that its frequency is qubit-state dependent. Measurement is performed by driving the resonator, where the transmitted resonator field yields information about the resonator frequency and thus the qubit state. Ideally, we could use arbitrarily strong resonator drives to achieve a target signal-to-noise ratio in the shortest possible time. However, experiments have shown that when the average resonator photon number exceeds a certain threshold, the qubit is excited out of its computational subspace in a process we refer to as a measurement-induced state transition (MIST). These transitions degrade readout fidelity, and constitute leakage which precludes further operation of the qubit in, for example, error correction. Here we study these transitions experimentally with a transmon qubit by measuring their dependence on qubit frequency, average resonator photon number, and qubit state, in the regime where the resonator frequency is lower than the qubit frequency. We observe signatures of resonant transitions between levels in the coupled qubit-resonator system that exhibit noisy behavior when measured repeatedly in time. We provide a semi-classical model of these transitions based on the rotating wave approximation and use it to predict the onset of state transitions in our experiments. Our results suggest the transmon is excited to levels near the top of its cosine potential following a state transition, where the charge dispersion of higher transmon levels explains the observed noisy behavior of state transitions. Moreover, we show that occupation in these higher energy levels poses a major challenge for fast qubit reset.

quant-ph↗
arXiv2023

Electric Vehicle Aggregation Review: Benefits and Vulnerabilities of Managing a Growing EV Fleet

Electric vehicles (EVs) are becoming more popular within the United States, making up an increasingly large portion of the US's electricity consumption. Hence, there is much attention has been directed on how to manage EVs within the power sector. A well-investigated strategy for managing the increase in electricity demand from EV charging is aggregation, which allows for an intermediary to manage electricity flow between EV owners and their utilities. When implemented effectively, EV aggregation provides key benefits to power grids by relieving electrical loads.. These benefits are aggregation's ability to shift EV loads to peak shave, which often leads to lower emissions, electricity generation prices, and consumer costs depending on the penetration levels of non-dispatchable electricity sources. This review seeks to appropriately highlight the broad vulnerabilities of EV aggregation alongside its benefits, namely those regarding battery degradation, rebound peaks, and cybersecurity. The holistic overview of EV aggregation provides comparisons that balance expectations with realistic performance.

eess.SY↗