arXiv subjects

Yongqiang Wang

Publications and source records attributed to Yongqiang Wang.

At least 55 records · Page 3Linked to original sources

arXiv2024

Tailoring Gradient Methods for Differentially-Private Distributed Optimization

Decentralized optimization is gaining increased traction due to its widespread applications in large-scale machine learning and multi-agent systems. The same mechanism that enables its success, i.e., information sharing among participating agents, however, also leads to the disclosure of individual agents' private information, which is unacceptable when sensitive data are involved. As differential privacy is becoming a de facto standard for privacy preservation, recently results have emerged integrating differential privacy with distributed optimization. However, directly incorporating differential privacy design in existing distributed optimization approaches significantly compromises optimization accuracy. In this paper, we propose to redesign and tailor gradient methods for differentially-private distributed optimization, and propose two differential-privacy oriented gradient methods that can ensure both rigorous epsilon-differential privacy and optimality. The first algorithm is based on static-consensus based gradient methods, and the second algorithm is based on dynamic-consensus (gradient-tracking) based distributed optimization methods and, hence, is applicable to general directed interaction graph topologies. Both algorithms can simultaneously ensure almost sure convergence to an optimal solution and a finite privacy budget, even when the number of iterations goes to infinity. To our knowledge, this is the first time that both goals are achieved simultaneously. Numerical simulations using a distributed estimation problem and experimental results on a benchmark dataset confirm the effectiveness of the proposed approaches.

math.OC↗

arXiv2024

Quantum Emitters in Aluminum Nitride Induced by Zirconium Ion Implantation

The integration of solid-state single-photon sources with foundry-compatible photonic platforms is crucial for practical and scalable quantum photonic applications. This study investigates aluminum nitride (AlN) as a material with properties highly suitable for integrated on-chip photonics specifically due to AlN capacity to host defect-center related single-photon emitters. We conduct a comprehensive study of the creation and photophysical properties of single-photon emitters in AlN utilizing Zirconium (Zr) and Krypton (Kr) heavy ion implantation and thermal annealing techniques. Guided by theoretical predictions, we assess the potential of Zr ions to create optically addressable spin-defects and employ Kr ions as an alternative approach that targets lattice defects without inducing chemical doping effects. With the 532 nm excitation wavelength, we found that single-photon emitters induced by ion implantation are primarily associated with vacancy-type defects in the AlN lattice for both Zr and Kr ions. The emitter density increases with the ion fluence, and there is an optimal value for the high density of emitters with low AlN background fluorescence. Additionally, under shorter excitation wavelength of 405 nm, Zr-implanted AlN exhibits isolated point-like emitters, which can be related to Zr-based defect complexes. This study provides important insights into the formation and properties of single-photon emitters in aluminum nitride induced by heavy ion implantation, contributing to the advancement of the aluminum nitride platform for on-chip quantum photonic applications.

physics.optics↗

arXiv2024

Multilingual and Fully Non-Autoregressive ASR with Large Language Model Fusion: A Comprehensive Study

In the era of large models, the autoregressive nature of decoding often results in latency serving as a significant bottleneck. We propose a non-autoregressive LM-fused ASR system that effectively leverages the parallelization capabilities of accelerator hardware. Our approach combines the Universal Speech Model (USM) and the PaLM 2 language model in per-segment scoring mode, achieving an average relative WER improvement across all languages of 10.8% on FLEURS and 3.6% on YouTube captioning. Furthermore, our comprehensive ablation study analyzes key parameters such as LLM size, context length, vocabulary size, fusion methodology. For instance, we explore the impact of LLM size ranging from 128M to 340B parameters on ASR performance. This study provides valuable insights into the factors influencing the effectiveness of practical large-scale LM-fused speech recognition systems.

cs.CL↗

arXiv2024

USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models

We introduce a multilingual speaker change detection model (USM-SCD) that can simultaneously detect speaker turns and perform ASR for 96 languages. This model is adapted from a speech foundation model trained on a large quantity of supervised and unsupervised data, demonstrating the utility of fine-tuning from a large generic foundation model for a downstream task. We analyze the performance of this multilingual speaker change detection model through a series of ablation studies. We show that the USM-SCD model can achieve more than 75% average speaker change detection F1 score across a test set that consists of data from 96 languages. On American English, the USM-SCD model can achieve an 85.8% speaker change detection F1 score across various public and internal test sets, beating the previous monolingual baseline model by 21% relative. We also show that we only need to fine-tune one-quarter of the trainable model parameters to achieve the best model performance. The USM-SCD model exhibits state-of-the-art ASR quality compared with a strong public ASR baseline, making it suitable to handle both tasks with negligible additional computational cost.

eess.AS↗

arXiv2023

Decentralized Gradient Methods with Time-varying Uncoordinated Stepsizes: Convergence Analysis and Privacy Design

Decentralized optimization enables a network of agents to cooperatively optimize an overall objective function without a central coordinator and is gaining increased attention in domains as diverse as control, sensor networks, data mining, and robotics. However, the information sharing among agents in decentralized optimization also discloses agents' information, which is undesirable or even unacceptable when involved data are sensitive. This paper proposes two gradient based decentralized optimization algorithms that can protect participating agents' privacy without compromising optimization accuracy or incurring heavy communication/computational overhead. This is in distinct difference from differential privacy based approaches which have to trade optimization accuracy for privacy, or encryption based approaches which incur heavy communication and computational overhead. Both algorithms leverage a judiciously designed mixing matrix and time-varying uncoordinated stepsizes to enable privacy, one using diminishing stepsizes while the other using non-diminishing stepsizes. In both algorithms, when interacting with any one of its neighbors, a participating agent only needs to share one message in each iteration to reach convergence to an exact optimal solution, which is in contrast to most gradient-tracking based algorithms requiring every agent to share two messages (an optimization variable and an auxiliary gradient-tracking variable) under non-diminishing stepsizes. Furthermore, both algorithms can guarantee the privacy of a participating agent even when all information shared by the agent are accessible to an adversary, a scenario in which most existing accuracy-maintaining privacy approaches will fail to protect privacy. Simulation results confirm the effectiveness of the proposed algorithms.

math.OC↗

arXiv2023

Locally Differentially Private Gradient Tracking for Distributed Online Learning over Directed Graphs

Distributed online learning has been proven extremely effective in solving large-scale machine learning problems over streaming data. However, information sharing between learners in distributed learning also raises concerns about the potential leakage of individual learners' sensitive data. To mitigate this risk, differential privacy, which is widely regarded as the "gold standard" for privacy protection, has been widely employed in many existing results on distributed online learning. However, these results often face a fundamental tradeoff between learning accuracy and privacy. In this paper, we propose a locally differentially private gradient tracking based distributed online learning algorithm that successfully circumvents this tradeoff. We prove that the proposed algorithm converges in mean square to the exact optimal solution while ensuring rigorous local differential privacy, with the cumulative privacy budget guaranteed to be finite even when the number of iterations tends to infinity. The algorithm is applicable even when the communication graph among learners is directed. To the best of our knowledge, this is the first result that simultaneously ensures learning accuracy and rigorous local differential privacy in distributed online learning over directed graphs. We evaluate our algorithm's performance by using multiple benchmark machine-learning applications, including logistic regression of the "Mushrooms" dataset and CNN-based image classification of the "MNIST" and "CIFAR-10" datasets, respectively. The experimental results confirm that the proposed algorithm outperforms existing counterparts in both training and testing accuracies.

cs.LG↗

arXiv2023

SLM: Bridge the thin gap between speech and text foundation models

We present a joint Speech and Language Model (SLM), a multitask, multilingual, and dual-modal model that takes advantage of pretrained foundational speech and language models. SLM freezes the pretrained foundation models to maximally preserves their capabilities, and only trains a simple adapter with just 1\% (156M) of the foundation models' parameters. This adaptation not only leads SLM to achieve strong performance on conventional tasks such as speech recognition (ASR) and speech translation (AST), but also introduces the novel capability of zero-shot instruction-following for more diverse tasks: given a speech input and a text instruction, SLM is able to perform unseen generation tasks including contextual biasing ASR using real-time context, dialog generation, speech continuation, and question answering, etc. Our approach demonstrates that the representational gap between pretrained speech and language models might be narrower than one would expect, and can be bridged by a simple adaptation mechanism. As a result, SLM is not only efficient to train, but also inherits strong capabilities already acquired in foundation models of different modalities.

cs.CL↗

arXiv2023

Google USM: Scaling Automatic Speech Recognition Beyond 100 Languages

We introduce the Universal Speech Model (USM), a single large model that performs automatic speech recognition (ASR) across 100+ languages. This is achieved by pre-training the encoder of the model on a large unlabeled multilingual dataset of 12 million (M) hours spanning over 300 languages, and fine-tuning on a smaller labeled dataset. We use multilingual pre-training with random-projection quantization and speech-text modality matching to achieve state-of-the-art performance on downstream multilingual ASR and speech-to-text translation tasks. We also demonstrate that despite using a labeled training set 1/7-th the size of that used for the Whisper model, our model exhibits comparable or better performance on both in-domain and out-of-domain speech recognition tasks across many languages.

cs.CL↗

arXiv2023

Undergraduate Research of Decentralized Localization of Roombas Through Usage of Wall-Finding Software

This paper introduces the research effort of an undergraduate research team in realizing robot localization. More specifically, the undergraduate research team developed and tested wall-following software that allowed a ground robot Roombas to independently find their positions within a defined space. The software also allows a robot to send its localized position to other Roombas, so that each Roomba knows its relative location to realize robot cooperation.

cs.RO↗

arXiv2023

Fast and High-Performance Learned Image Compression With Improved Checkerboard Context Model, Deformable Residual Module, and Knowledge Distillation

Deep learning-based image compression has made great progresses recently. However, many leading schemes use serial context-adaptive entropy model to improve the rate-distortion (R-D) performance, which is very slow. In addition, the complexities of the encoding and decoding networks are quite high and not suitable for many practical applications. In this paper, we introduce four techniques to balance the trade-off between the complexity and performance. We are the first to introduce deformable convolutional module in compression framework, which can remove more redundancies in the input image, thereby enhancing compression performance. Second, we design a checkerboard context model with two separate distribution parameter estimation networks and different probability models, which enables parallel decoding without sacrificing the performance compared to the sequential context-adaptive model. Third, we develop an improved three-step knowledge distillation and training scheme to achieve different trade-offs between the complexity and the performance of the decoder network, which transfers both the final and intermediate results of the teacher network to the student network to help its training. Fourth, we introduce $L_{1}$ regularization to make the numerical values of the latent representation more sparse. Then we only encode non-zero channels in the encoding and decoding process, which can greatly reduce the encoding and decoding time. Experiments show that compared to the state-of-the-art learned image coding scheme, our method can be about 20 times faster in encoding and 70-90 times faster in decoding, and our R-D performance is also $2.3 \%$ higher. Our method outperforms the traditional approach in H.266/VVC-intra (4:4:4) and some leading learned schemes in terms of PSNR and MS-SSIM metrics when testing on Kodak and Tecnick-40 datasets.

eess.IV↗

arXiv2023

Enhanced Residual SwinV2 Transformer for Learned Image Compression

Recently, the deep learning technology has been successfully applied in the field of image compression, leading to superior rate-distortion performance. However, a challenge of many learning-based approaches is that they often achieve better performance via sacrificing complexity, which making practical deployment difficult. To alleviate this issue, in this paper, we propose an effective and efficient learned image compression framework based on an enhanced residual Swinv2 transformer. To enhance the nonlinear representation of images in our framework, we use a feature enhancement module that consists of three consecutive convolutional layers. In the subsequent coding and hyper coding steps, we utilize a SwinV2 transformer-based attention mechanism to process the input image. The SwinV2 model can help to reduce model complexity while maintaining high performance. Experimental results show that the proposed method achieves comparable performance compared to some recent learned image compression methods on Kodak and Tecnick datasets, and outperforms some traditional codecs including VVC. In particular, our method achieves comparable results while reducing model complexity by 56% compared to these recent methods.

eess.IV↗

arXiv2023

Microvasculature Segmentation in Human BioMolecular Atlas Program (HuBMAP)

Image segmentation serves as a critical tool across a range of applications, encompassing autonomous driving's pedestrian detection and pre-operative tumor delineation in the medical sector. Among these applications, we focus on the National Institutes of Health's (NIH) Human BioMolecular Atlas Program (HuBMAP), a significant initiative aimed at creating detailed cellular maps of the human body. In this study, we concentrate on segmenting various microvascular structures in human kidneys, utilizing 2D Periodic Acid-Schiff (PAS)-stained histology images. Our methodology begins with a foundational FastAI U-Net model, upon which we investigate alternative backbone architectures, delve into deeper models, and experiment with Feature Pyramid Networks. We rigorously evaluate these varied approaches by benchmarking their performance against our baseline U-Net model. This study thus offers a comprehensive exploration of cutting-edge segmentation techniques, providing valuable insights for future research in the field.

eess.IV↗

arXiv2023

MFMAN-YOLO: A Method for Detecting Pole-like Obstacles in Complex Environment

In real-world traffic, there are various uncertainties and complexities in road and weather conditions. To solve the problem that the feature information of pole-like obstacles in complex environments is easily lost, resulting in low detection accuracy and low real-time performance, a multi-scale hybrid attention mechanism detection algorithm is proposed in this paper. First, the optimal transport function Monge-Kantorovich (MK) is incorporated not only to solve the problem of overlapping multiple prediction frames with optimal matching but also the MK function can be regularized to prevent model over-fitting; then, the features at different scales are up-sampled separately according to the optimized efficient multi-scale feature pyramid. Finally, the extraction of multi-scale feature space channel information is enhanced in complex environments based on the hybrid attention mechanism, which suppresses the irrelevant complex environment background information and focuses the feature information of pole-like obstacles. Meanwhile, this paper conducts real road test experiments in a variety of complex environments. The experimental results show that the detection precision, recall, and average precision of the method are 94.7%, 93.1%, and 97.4%, respectively, and the detection frame rate is 400 f/s. This research method can detect pole-like obstacles in a complex road environment in real time and accurately, which further promotes innovation and progress in the field of automatic driving.

cs.CV↗

arXiv2023

A Robust Dynamic Average Consensus Algorithm that Ensures both Differential Privacy and Accurate Convergence

We propose a new dynamic average consensus algorithm that is robust to information-sharing noise arising from differential-privacy design. Not only is dynamic average consensus widely used in cooperative control and distributed tracking, it is also a fundamental building block in numerous distributed computation algorithms such as multi-agent optimization and distributed Nash equilibrium seeking. We propose a new dynamic average consensus algorithm that is robust to persistent and independent information-sharing noise added for the purpose of differential-privacy protection. In fact, the algorithm can ensure both provable convergence to the exact average reference signal and rigorous epsilon-differential privacy (even when the number of iterations tends to infinity), which, to our knowledge, has not been achieved before in average consensus algorithms. Given that channel noise in communication can be viewed as a special case of differential-privacy noise, the algorithm can also be used to counteract communication imperfections. Numerical simulation results confirm the effectiveness of the proposed approach.

cs.CR↗

arXiv2023

Differentially-private Distributed Algorithms for Aggregative Games with Guaranteed Convergence

The distributed computation of a Nash equilibrium in aggregative games is gaining increased traction in recent years. Of particular interest is the mediator-free scenario where individual players only access or observe the decisions of their neighbors due to practical constraints. Given the competitive rivalry among participating players, protecting the privacy of individual players becomes imperative when sensitive information is involved. We propose a fully distributed equilibrium-computation approach for aggregative games that can achieve both rigorous differential privacy and guaranteed computation accuracy of the Nash equilibrium. This is in sharp contrast to existing differential-privacy solutions for aggregative games that have to either sacrifice the accuracy of equilibrium computation to gain rigorous privacy guarantees, or allow the cumulative privacy budget to grow unbounded, hence losing privacy guarantees, as iteration proceeds. Our approach uses independent noises across players, thus making it effective even when adversaries have access to all shared messages as well as the underlying algorithm structure. The encryption-free nature of the proposed approach, also ensures efficiency in computation and communication. The approach is also applicable in stochastic aggregative games, able to ensure both rigorous differential privacy and guaranteed computation accuracy of the Nash equilibrium when individual players only have stochastic estimates of their pseudo-gradient mappings. Numerical comparisons with existing counterparts confirm the effectiveness of the proposed approach.

cs.GT↗

arXiv2023

AudioPaLM: A Large Language Model That Can Speak and Listen

We introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt. We release examples of our method at https://google-research.github.io/seanet/audiopalm/examples

cs.CL↗

arXiv2023

Gradient-tracking Based Differentially Private Distributed Optimization with Enhanced Optimization Accuracy

Privacy protection has become an increasingly pressing requirement in distributed optimization. However, equipping distributed optimization with differential privacy, the state-of-the-art privacy protection mechanism, will unavoidably compromise optimization accuracy. In this paper, we propose an algorithm to achieve rigorous $ε$-differential privacy in gradient-tracking based distributed optimization with enhanced optimization accuracy. More specifically, to suppress the influence of differential-privacy noise, we propose a new robust gradient-tracking based distributed optimization algorithm that allows both stepsize and the variance of injected noise to vary with time. Then, we establish a new analyzing approach that can characterize the convergence of the gradient-tracking based algorithm under both constant and time-varying stespsizes. To our knowledge, this is the first analyzing framework that can treat gradient-tracking based distributed optimization under both constant and time-varying stepsizes in a unified manner. More importantly, the new analyzing approach gives a much less conservative analytical bound on the stepsize compared with existing proof techniques for gradient-tracking based distributed optimization. We also theoretically characterize the influence of differential-privacy design on the accuracy of distributed optimization, which reveals that inter-agent interaction has a significant impact on the final optimization accuracy. Numerical simulation results confirm the theoretical predictions.

math.OC↗

arXiv2023

The spectral radius of a square matrix can be approximated by its "weighted" spectral norm

In distributed optimization or Nash-equilibrium seeking over directed graphs, it is crucial to find a matrix norm under which the disagreement of individual agents' states contracts. In existing results, the matrix norm is usually defined by approximating the spectral radius of the matrix, which is possible when the matrix is real and has zero row-sums. In this brief note, we show that this technique can be applied to general square complex matrices. More specifically, we prove that the spectral radius of any complex square matrix can be approximated by its "weighted" spectral norm to an arbitrary degree of accuracy.

math.OC↗