arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 1,063 records · Page 59Linked to original sources

arXiv2023

Geometric phases of mixed quantum states: A comparative study of interferometric and Uhlmann phases

Two geometric phases of mixed quantum states, known as the interferometric phase and Uhlmann phase, are generalizations of the Berry phase of pure states. After reviewing the two geometric phases and examining their parallel-transport conditions, we specify a class of cyclic processes that are compatible with both conditions and therefore accumulate both phases through their definitions, respectively. Those processes then facilitate a fair comparison between the two phases. We present exact solutions of two-level and three-level systems to contrast the two phases. While the interferometric phase exhibits finite-temperature transitions only in the three-level system but not the two-level system, the Uhlmann phase shows finite-temperature transitions in both cases. Thus, using the two geometric phases as finite-temperature topological indicators demonstrates the rich physics of topology of mixed states.

quant-ph↗

arXiv2023

Decentralized Federated Learning via MIMO Over-the-Air Computation: Consensus Analysis and Performance Optimization

Decentralized federated learning (DFL), inherited from distributed optimization, is an emerging paradigm to leverage the explosively growing data from wireless devices in a fully distributed manner.DFL enables joint training of machine learning model under device to device (D2D) communication fashion without the coordination of a parameter server. However, the deployment of wireless DFL is facing some pivotal challenges. Communication is a critical bottleneck due to the required extensive message exchange between neighbor devices to share the learned model. Besides, consensus becomes increasingly difficult as the number of devices grows because there is no available central server to perform coordination. To overcome these difficulties, this paper proposes employing over-the-air computation (Aircomp) to improve communication efficiency by exploiting the superposition property of analog waveform in multi-access channels, and introduce the mixing matrix mechanism to promote consensus using the spectral property of symmetric doubly stochastic matrix. Specifically, we develop a novel multiple-input multiple-output over-the-air DFL (MIMO OA-DFL) framework to study over-the-air DFL problem over MIMO multiple access channels. We conduct a general convergence analysis to quantitatively capture the influence of aggregation weight and communication error on the MIMO OA-DFL performance in \emph{ad hoc} networks. The result shows that the communication error together with the spectral gap of mixing matrix has a significant impact on the learning performance. Based on this, a joint communication-learning optimization problem is formulated to optimize transceiver beamformers and mixing matrix. Extensive numerical experiments are performed to reveal the characteristics of different topologies and demonstrate the substantial learning performance enhancement of our proposed algorithm.

eess.SP↗

arXiv2023

D-type Minimal Conformal Matter: Quantum Curves, Elliptic Garnier Systems, and the 5d Descendants

We study the quantization of the 6d Seiberg-Witten curve for D-type minimal conformal matter theories compactified on a two-torus. The quantized 6d curve turns out to be a difference equation established via introducing codimension two and four surface defects. We show that, in the Nekrasov-Shatashvili limit, the 6d partition function with insertions of codimension two and four defects serve as the eigenfunction and eigenvalues of the difference equation, respectively. We further identify the quantum curve of D-type minimal conformal matters with an elliptic Garnier system recently studied in the integrability community. At last, as a concrete consequence of our elliptic quantum curve, we study its RG flows to obtain various quantum curves of 5d ${\rm Sp}(N)+N_f \mathsf{F},N_f\leq 2N+5$ theories.

hep-th↗

arXiv2023

Alternating Updates for Efficient Transformers

It has been well established that increasing scale in deep transformer networks leads to improved quality and performance. However, this increase in scale often comes with prohibitive increases in compute cost and inference latency. We introduce Alternating Updates (AltUp), a simple-to-implement method to increase a model's capacity without the computational burden. AltUp enables the widening of the learned representation, i.e., the token embedding, while only incurring a negligible increase in latency. AltUp achieves this by working on a subblock of the widened representation at each layer and using a predict-and-correct mechanism to update the inactivated blocks. We present extensions of AltUp, such as its applicability to the sequence dimension, and demonstrate how AltUp can be synergistically combined with existing approaches, such as Sparse Mixture-of-Experts models, to obtain efficient models with even higher capacity. Our experiments on benchmark transformer models and language tasks demonstrate the consistent effectiveness of AltUp on a diverse set of scenarios. Notably, on SuperGLUE and SQuAD benchmarks, AltUp enables up to $87\%$ speedup relative to the dense baselines at the same accuracy.

cs.LG↗

arXiv2023

Textbooks Are All You Need

We introduce phi-1, a new large language model for code, with significantly smaller size than competing models: phi-1 is a Transformer-based model with 1.3B parameters, trained for 4 days on 8 A100s, using a selection of ``textbook quality" data from the web (6B tokens) and synthetically generated textbooks and exercises with GPT-3.5 (1B tokens). Despite this small scale, phi-1 attains pass@1 accuracy 50.6% on HumanEval and 55.5% on MBPP. It also displays surprising emergent properties compared to phi-1-base, our model before our finetuning stage on a dataset of coding exercises, and phi-1-small, a smaller model with 350M parameters trained with the same pipeline as phi-1 that still achieves 45% on HumanEval.

cs.CL↗

arXiv2023

UV-Bright Star-Forming Clumps and Their Host Galaxies in UVCANDELS at 0.5 $\leq$ z $\leq$ 1

Giant star-forming clumps are a prominent feature of star-forming galaxies (SFGs) and contain important clues on galaxy formation and evolution. However, basic demographics of clumps and their host galaxies remain uncertain. Using the HST/WFC3 F275W images from the Ultraviolet Imaging of the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey (UVCANDELS), we detect and analyze giant star-forming clumps in galaxies at 0.5 $\leq$ z $\leq$ 1, connecting two epochs when clumps are common (at cosmic high-noon, z $\sim$ 2) and rare (in the local universe). We construct a clump sample whose rest-frame 1600 Å luminosity is 3 times higher than the most luminous local HII regions (M$_{UV} \leq -$16 AB). In our sample, 35 $\pm$ 3$\%$ of low-mass galaxies (log[M$_{*}$/M$_{\odot}$] $<$ 10) are clumpy (i.e., containing at least one off-center clump). This fraction changes to 22 $\pm$ 3$\%$ and 22 $\pm$ 4$\%$ for intermediate (10 $\leq$ log[M$_{*}$/M$_{\odot}$] $\leq$ 10.5) and high-mass (log[M$_{*}$/M$_{\odot}$] $>$ 10.5) galaxies in agreement with previous studies. When compared to similar-mass non-clumpy SFGs, low- and intermediate-mass clumpy SFGs tend to have higher SFRs and bluer rest-frame U-V colors, while high-mass clumpy SFGs tend to be larger than non-clumpy SFGs. However, clumpy and non-clumpy SFGs have similar Sérsic index, indicating a similar underlying density profile. Furthermore, we investigate how UV luminosity of star-forming regions correlates with the physical properties of host galaxies. On average, more luminous star-forming regions reside in more luminous, smaller, and/or higher-specific SFR galaxies and are found closer to their hosts' galactic center.

astro-ph.GA↗

arXiv2023

Controlling Neural Style Transfer with Deep Reinforcement Learning

Controlling the degree of stylization in the Neural Style Transfer (NST) is a little tricky since it usually needs hand-engineering on hyper-parameters. In this paper, we propose the first deep Reinforcement Learning (RL) based architecture that splits one-step style transfer into a step-wise process for the NST task. Our RL-based method tends to preserve more details and structures of the content image in early steps, and synthesize more style patterns in later steps. It is a user-easily-controlled style-transfer method. Additionally, as our RL-based model performs the stylization progressively, it is lightweight and has lower computational complexity than existing one-step Deep Learning (DL) based models. Experimental results demonstrate the effectiveness and robustness of our method.

cs.CV↗

arXiv2023

Unlocking the Potential of Deep Learning in Peak-Hour Series Forecasting

Unlocking the potential of deep learning in Peak-Hour Series Forecasting (PHSF) remains a critical yet underexplored task in various domains. While state-of-the-art deep learning models excel in regular Time Series Forecasting (TSF), they struggle to achieve comparable results in PHSF. This can be attributed to the challenges posed by the high degree of non-stationarity in peak-hour series, which makes direct forecasting more difficult than standard TSF. Additionally, manually extracting the maximum value from regular forecasting results leads to suboptimal performance due to models minimizing the mean deficit. To address these issues, this paper presents Seq2Peak, a novel framework designed specifically for PHSF tasks, bridging the performance gap observed in TSF models. Seq2Peak offers two key components: the CyclicNorm pipeline to mitigate the non-stationarity issue and a simple yet effective trainable-parameter-free peak-hour decoder with a hybrid loss function that utilizes both the original series and peak-hour series as supervised signals. Extensive experimentation on publicly available time series datasets demonstrates the effectiveness of the proposed framework, yielding a remarkable average relative improvement of 37.7% across four real-world datasets for both transformer- and non-transformer-based TSF models.

cs.LG↗

arXiv2023

HoloAssist: an Egocentric Human Interaction Dataset for Interactive AI Assistants in the Real World

Building an interactive AI assistant that can perceive, reason, and collaborate with humans in the real world has been a long-standing pursuit in the AI community. This work is part of a broader research effort to develop intelligent agents that can interactively guide humans through performing tasks in the physical world. As a first step in this direction, we introduce HoloAssist, a large-scale egocentric human interaction dataset, where two people collaboratively complete physical manipulation tasks. The task performer executes the task while wearing a mixed-reality headset that captures seven synchronized data streams. The task instructor watches the performer's egocentric video in real time and guides them verbally. By augmenting the data with action and conversational annotations and observing the rich behaviors of various participants, we present key insights into how human assistants correct mistakes, intervene in the task completion procedure, and ground their instructions to the environment. HoloAssist spans 166 hours of data captured by 350 unique instructor-performer pairs. Furthermore, we construct and present benchmarks on mistake detection, intervention type prediction, and hand forecasting, along with detailed analysis. We expect HoloAssist will provide an important resource for building AI assistants that can fluidly collaborate with humans in the real world. Data can be downloaded at https://holoassist.github.io/.

cs.CV↗

arXiv2023

Quantum Self-Attention Neural Networks for Text Classification

An emerging direction of quantum computing is to establish meaningful quantum applications in various fields of artificial intelligence, including natural language processing (NLP). Although some efforts based on syntactic analysis have opened the door to research in Quantum NLP (QNLP), limitations such as heavy syntactic preprocessing and syntax-dependent network architecture make them impracticable on larger and real-world data sets. In this paper, we propose a new simple network architecture, called the quantum self-attention neural network (QSANN), which can compensate for these limitations. Specifically, we introduce the self-attention mechanism into quantum neural networks and then utilize a Gaussian projected quantum self-attention serving as a sensible quantum version of self-attention. As a result, QSANN is effective and scalable on larger data sets and has the desirable property of being implementable on near-term quantum devices. In particular, our QSANN outperforms the best existing QNLP model based on syntactic analysis as well as a simple classical self-attention neural network in numerical experiments of text classification tasks on public data sets. We further show that our method exhibits robustness to low-level quantum noises and showcases resilience to quantum neural network architectures.

quant-ph↗

arXiv2023

Statistical Analysis of Quantum State Learning Process in Quantum Neural Networks

Quantum neural networks (QNNs) have been a promising framework in pursuing near-term quantum advantage in various fields, where many applications can be viewed as learning a quantum state that encodes useful data. As a quantum analog of probability distribution learning, quantum state learning is theoretically and practically essential in quantum machine learning. In this paper, we develop a no-go theorem for learning an unknown quantum state with QNNs even starting from a high-fidelity initial state. We prove that when the loss value is lower than a critical threshold, the probability of avoiding local minima vanishes exponentially with the qubit count, while only grows polynomially with the circuit depth. The curvature of local minima is concentrated to the quantum Fisher information times a loss-dependent constant, which characterizes the sensibility of the output state with respect to parameters in QNNs. These results hold for any circuit structures, initialization strategies, and work for both fixed ansatzes and adaptive methods. Extensive numerical simulations are performed to validate our theoretical results. Our findings place generic limits on good initial guesses and adaptive methods for improving the learnability and scalability of QNNs, and deepen the understanding of prior information's role in QNNs.

quant-ph↗

arXiv2023

WiCV@CVPR2023: The Eleventh Women In Computer Vision Workshop at the Annual CVPR Conference

In this paper, we present the details of Women in Computer Vision Workshop - WiCV 2023, organized alongside the hybrid CVPR 2023 in Vancouver, Canada. WiCV aims to amplify the voices of underrepresented women in the computer vision community, fostering increased visibility in both academia and industry. We believe that such events play a vital role in addressing gender imbalances within the field. The annual WiCV@CVPR workshop offers a) opportunity for collaboration between researchers from minority groups, b) mentorship for female junior researchers, c) financial support to presenters to alleviate finanacial burdens and d) a diverse array of role models who can inspire younger researchers at the outset of their careers. In this paper, we present a comprehensive report on the workshop program, historical trends from the past WiCV@CVPR events, and a summary of statistics related to presenters, attendees, and sponsorship for the WiCV 2023 workshop.

cs.CV↗

arXiv2023

Optimal Beamforming for MIMO DFRC Systems with Transmit Covariance Constraints

This paper optimizes the beamforming design of a downlink multiple-input multiple-output (MIMO) dual-function radar-communication (DFRC) system to maximize the weighted communication sum-rate under a prescribed transmit covariance constraint for radar performance guarantee. In the single-user case, we show that the transmit covariance constraint implies the existence of inherent orthogonality among the transmit beamforming vectors in use. Then, leveraging Cauchy's interlace theorem, we derive the globally optimal transmit and receive beamforming solution in closed form. In the multi-user case, we exploit the connection between the weighted sum-rate and weighted minimum mean squared error (MMSE) to reformulate the intended problem, and develop a block-coordinate-descent (BCD) algorithm to iteratively compute the transmit beamforming and receive beamforming solutions. Under this approach, we reveal that the optimal receive beamforming is given by the classic MMSE one and the optimal transmit beamforming design amounts to solving an orthogonal Procrustes problem, thereby allowing for closed-form solutions to subproblems in each BCD step and fast convergence of the proposed algorithm to a high-quality (near-optimal) overall beamforming design. Numerical results demonstrate the superiority of our approach to the existing methods, with at least 40% higher sum-rate under a multi-user MIMO setting in the high signal-to-noise regime.

eess.SP↗

arXiv2023

Arbitrary Reduction of MRI Inter-slice Spacing Using Hierarchical Feature Conditional Diffusion

Magnetic resonance (MR) images collected in 2D scanning protocols typically have large inter-slice spacing, resulting in high in-plane resolution but reduced through-plane resolution. Super-resolution techniques can reduce the inter-slice spacing of 2D scanned MR images, facilitating the downstream visual experience and computer-aided diagnosis. However, most existing super-resolution methods are trained at a fixed scaling ratio, which is inconvenient in clinical settings where MR scanning may have varying inter-slice spacings. To solve this issue, we propose Hierarchical Feature Conditional Diffusion (HiFi-Diff)} for arbitrary reduction of MR inter-slice spacing. Given two adjacent MR slices and the relative positional offset, HiFi-Diff can iteratively convert a Gaussian noise map into any desired in-between MR slice. Furthermore, to enable fine-grained conditioning, the Hierarchical Feature Extraction (HiFE) module is proposed to hierarchically extract conditional features and conduct element-wise modulation. Our experimental results on the publicly available HCP-1200 dataset demonstrate the high-fidelity super-resolution capability of HiFi-Diff and its efficacy in enhancing downstream segmentation performance.

eess.IV↗

arXiv2023

For A More Comprehensive Evaluation of 6DoF Object Pose Tracking

Previous evaluations on 6DoF object pose tracking have presented obvious limitations along with the development of this area. In particular, the evaluation protocols are not unified for different methods, the widely-used YCBV dataset contains significant annotation error, and the error metrics also may be biased. As a result, it is hard to fairly compare the methods, which has became a big obstacle for developing new algorithms. In this paper we contribute a unified benchmark to address the above problems. For more accurate annotation of YCBV, we propose a multi-view multi-object global pose refinement method, which can jointly refine the poses of all objects and view cameras, resulting in sub-pixel sub-millimeter alignment errors. The limitations of previous scoring methods and error metrics are analyzed, based on which we introduce our improved evaluation methods. The unified benchmark takes both YCBV and BCOT as base datasets, which are shown to be complementary in scene categories. In experiments, we validate the precision and reliability of the proposed global pose refinement method with a realistic semi-synthesized dataset particularly for YCBV, and then present the benchmark results unifying learning&non-learning and RGB&RGBD methods, with some finds not discovered in previous studies.

cs.CV↗

arXiv2023

DDSP-based Neural Waveform Synthesis of Polyphonic Guitar Performance from String-wise MIDI Input

We explore the use of neural synthesis for acoustic guitar from string-wise MIDI input. We propose four different systems and compare them with both objective metrics and subjective evaluation against natural audio and a sample-based baseline. We iteratively develop these four systems by making various considerations on the architecture and intermediate tasks, such as predicting pitch and loudness control features. We find that formulating the control feature prediction task as a classification task rather than a regression task yields better results. Furthermore, we find that our simplest proposed system, which directly predicts synthesis parameters from MIDI input performs the best out of the four proposed systems. Audio examples are available at https://erl-j.github.io/neural-guitar-web-supplement.

cs.SD↗

arXiv2023

Speaker anonymization using orthogonal Householder neural network

Speaker anonymization aims to conceal a speaker's identity while preserving content information in speech. Current mainstream neural-network speaker anonymization systems disentangle speech into prosody-related, content, and speaker representations. The speaker representation is then anonymized by a selection-based speaker anonymizer that uses a mean vector over a set of randomly selected speaker vectors from an external pool of English speakers. However, the resulting anonymized vectors are subject to severe privacy leakage against powerful attackers, reduction in speaker diversity, and language mismatch problems for unseen-language speaker anonymization. To generate diverse, language-neutral speaker vectors, this paper proposes an anonymizer based on an orthogonal Householder neural network (OHNN). Specifically, the OHNN acts like a rotation to transform the original speaker vectors into anonymized speaker vectors, which are constrained to follow the distribution over the original speaker vector space. A basic classification loss is introduced to ensure that anonymized speaker vectors from different speakers have unique speaker identities. To further protect speaker identities, an improved classification loss and similarity loss are used to push original-anonymized sample pairs away from each other. Experiments on VoicePrivacy Challenge datasets in English and the \textit{AISHELL-3} dataset in Mandarin demonstrate the proposed anonymizer's effectiveness.

cs.SD↗

arXiv2023

SynVox2: Towards a privacy-friendly VoxCeleb2 dataset

The success of deep learning in speaker recognition relies heavily on the use of large datasets. However, the data-hungry nature of deep learning methods has already being questioned on account the ethical, privacy, and legal concerns that arise when using large-scale datasets of natural speech collected from real human speakers. For example, the widely-used VoxCeleb2 dataset for speaker recognition is no longer accessible from the official website. To mitigate these concerns, this work presents an initiative to generate a privacy-friendly synthetic VoxCeleb2 dataset that ensures the quality of the generated speech in terms of privacy, utility, and fairness. We also discuss the challenges of using synthetic data for the downstream task of speaker verification.

cs.SD↗