Search arXiv⌕ Search

arXiv subjects

Xiang Li

Publications and source records attributed to Xiang Li.

At least 1,009 records · Page 56Linked to original sources
arXiv2023

Shapley Value on Probabilistic Classifiers

Data valuation has become an increasingly significant discipline in data science due to the economic value of data. In the context of machine learning (ML), data valuation methods aim to equitably measure the contribution of each data point to the utility of an ML model. One prevalent method is Shapley value, which helps identify data points that are beneficial or detrimental to an ML model. However, traditional Shapley-based data valuation methods may not effectively distinguish between beneficial and detrimental training data points for probabilistic classifiers. In this paper, we propose Probabilistic Shapley (P-Shapley) value by constructing a probability-wise utility function that leverages the predicted class probabilities of probabilistic classifiers rather than binarized prediction results in the traditional Shapley value. We also offer several activation functions for confidence calibration to effectively quantify the marginal contribution of each data point to the probabilistic classifiers. Extensive experiments on four real-world datasets demonstrate the effectiveness of our proposed P-Shapley value in evaluating the importance of data for building a high-usability and trustworthy ML model.

cs.LG↗
arXiv2023

Tunable Coupling Architectures with Capacitively Connecting Pads for Large-Scale Superconducting Multi-Qubit Processors

We have proposed and experimentally verified a tunable inter-qubit coupling scheme for large-scale integration of superconducting qubits. The key feature of the scheme is the insertion of connecting pads between qubit and tunable coupling element. In such a way, the distance between two qubits can be increased considerably to a few millimeters, leaving enough space for arranging control lines, readout resonators and other necessary structures. The increased inter-qubit distance provides more wiring space for flip-chip process and reduces crosstalk between qubits and from control lines to qubits. We use the term Tunable Coupler with Capacitively Connecting Pad (TCCP) to name the tunable coupling part that consists of a transmon coupler and capacitively connecting pads. With the different placement of connecting pads, different TCCP architectures can be realized. We have designed and fabricated a few multi-qubit devices in which TCCP is used for coupling. The measured results show that the performance of the qubits coupled by the TCCP, such as $T_1$ and $T_2$, was similar to that of the traditional transmon qubits without TCCP. Meanwhile, our TCCP also exhibited a wide tunable range of the effective coupling strength and a low residual ZZ interaction between the qubits by properly tuning the parameters on the design. Finally, we successfully implemented an adiabatic CZ gate with TCCP. Furthermore, by introducing TCCP, we also discuss the realization of the flip-chip process and tunable coupling qubits between different chips.

quant-ph↗
arXiv2023

PaintSeg: Training-free Segmentation via Painting

The paper introduces PaintSeg, a new unsupervised method for segmenting objects without any training. We propose an adversarial masked contrastive painting (AMCP) process, which creates a contrast between the original image and a painted image in which a masked area is painted using off-the-shelf generative models. During the painting process, inpainting and outpainting are alternated, with the former masking the foreground and filling in the background, and the latter masking the background while recovering the missing part of the foreground object. Inpainting and outpainting, also referred to as I-step and O-step, allow our method to gradually advance the target segmentation mask toward the ground truth without supervision or training. PaintSeg can be configured to work with a variety of prompts, e.g. coarse masks, boxes, scribbles, and points. Our experimental results demonstrate that PaintSeg outperforms existing approaches in coarse mask-prompt, box-prompt, and point-prompt segmentation tasks, providing a training-free solution suitable for unsupervised segmentation.

cs.CV↗
arXiv2023

Exploring Better Text Image Translation with Multimodal Codebook

Text image translation (TIT) aims to translate the source texts embedded in the image to target translations, which has a wide range of applications and thus has important research value. However, current studies on TIT are confronted with two main bottlenecks: 1) this task lacks a publicly available TIT dataset, 2) dominant models are constructed in a cascaded manner, which tends to suffer from the error propagation of optical character recognition (OCR). In this work, we first annotate a Chinese-English TIT dataset named OCRMT30K, providing convenience for subsequent studies. Then, we propose a TIT model with a multimodal codebook, which is able to associate the image with relevant texts, providing useful supplementary information for translation. Moreover, we present a multi-stage training framework involving text machine translation, image-text alignment, and TIT tasks, which fully exploits additional bilingual texts, OCR dataset and our OCRMT30K dataset to train our model. Extensive experiments and in-depth analyses strongly demonstrate the effectiveness of our proposed model and training framework.

cs.CL↗
arXiv2023

Video ChatCaptioner: Towards Enriched Spatiotemporal Descriptions

Video captioning aims to convey dynamic scenes from videos using natural language, facilitating the understanding of spatiotemporal information within our environment. Although there have been recent advances, generating detailed and enriched video descriptions continues to be a substantial challenge. In this work, we introduce Video ChatCaptioner, an innovative approach for creating more comprehensive spatiotemporal video descriptions. Our method employs a ChatGPT model as a controller, specifically designed to select frames for posing video content-driven questions. Subsequently, a robust algorithm is utilized to answer these visual queries. This question-answer framework effectively uncovers intricate video details and shows promise as a method for enhancing video content. Following multiple conversational rounds, ChatGPT can summarize enriched video content based on previous conversations. We qualitatively demonstrate that our Video ChatCaptioner can generate captions containing more visual details about the videos. The code is publicly available at https://github.com/Vision-CAIR/ChatCaptioner

cs.CV↗
arXiv2023

Is Synthetic Data From Diffusion Models Ready for Knowledge Distillation?

Diffusion models have recently achieved astonishing performance in generating high-fidelity photo-realistic images. Given their huge success, it is still unclear whether synthetic images are applicable for knowledge distillation when real images are unavailable. In this paper, we extensively study whether and how synthetic images produced from state-of-the-art diffusion models can be used for knowledge distillation without access to real images, and obtain three key conclusions: (1) synthetic data from diffusion models can easily lead to state-of-the-art performance among existing synthesis-based distillation methods, (2) low-fidelity synthetic images are better teaching materials, and (3) relatively weak classifiers are better teachers. Code is available at https://github.com/zhengli97/DM-KD.

cs.CV↗
arXiv2023

Two Sides of One Coin: the Limits of Untuned SGD and the Power of Adaptive Methods

The classical analysis of Stochastic Gradient Descent (SGD) with polynomially decaying stepsize $η_t = η/\sqrt{t}$ relies on well-tuned $η$ depending on problem parameters such as Lipschitz smoothness constant, which is often unknown in practice. In this work, we prove that SGD with arbitrary $η> 0$, referred to as untuned SGD, still attains an order-optimal convergence rate $\widetilde{O}(T^{-1/4})$ in terms of gradient norm for minimizing smooth objectives. Unfortunately, it comes at the expense of a catastrophic exponential dependence on the smoothness constant, which we show is unavoidable for this scheme even in the noiseless setting. We then examine three families of adaptive methods $\unicode{x2013}$ Normalized SGD (NSGD), AMSGrad, and AdaGrad $\unicode{x2013}$ unveiling their power in preventing such exponential dependency in the absence of information about the smoothness parameter and boundedness of stochastic gradients. Our results provide theoretical justification for the advantage of adaptive methods over untuned SGD in alleviating the issue with large gradients.

math.OC↗
arXiv2023

Ranking-Enhanced Unsupervised Sentence Representation Learning

Unsupervised sentence representation learning has progressed through contrastive learning and data augmentation methods such as dropout masking. Despite this progress, sentence encoders are still limited to using only an input sentence when predicting its semantic vector. In this work, we show that the semantic meaning of a sentence is also determined by nearest-neighbor sentences that are similar to the input sentence. Based on this finding, we propose a novel unsupervised sentence encoder, RankEncoder. RankEncoder predicts the semantic vector of an input sentence by leveraging its relationship with other sentences in an external corpus, as well as the input sentence itself. We evaluate RankEncoder on semantic textual benchmark datasets. From the experimental results, we verify that 1) RankEncoder achieves 80.07% Spearman's correlation, a 1.1% absolute improvement compared to the previous state-of-the-art performance, 2) RankEncoder is universally applicable to existing unsupervised sentence embedding methods, and 3) RankEncoder is specifically effective for predicting the similarity scores of similar sentence pairs.

cs.CL↗
arXiv2023

When Gradient Descent Meets Derivative-Free Optimization: A Match Made in Black-Box Scenario

Large pre-trained language models (PLMs) have garnered significant attention for their versatility and potential for solving a wide spectrum of natural language processing (NLP) tasks. However, the cost of running these PLMs may be prohibitive. Furthermore, PLMs may not be open-sourced due to commercial considerations and potential risks of misuse, such as GPT-3. The parameters and gradients of PLMs are unavailable in this scenario. To solve the issue, black-box tuning has been proposed, which utilizes derivative-free optimization (DFO), instead of gradient descent, for training task-specific continuous prompts. However, these gradient-free methods still exhibit a significant gap compared to gradient-based methods. In this paper, we introduce gradient descent into black-box tuning scenario through knowledge distillation. Furthermore, we propose a novel method GDFO, which integrates gradient descent and derivative-free optimization to optimize task-specific continuous prompts in a harmonized manner. Experimental results show that GDFO can achieve significant performance gains over previous state-of-the-art methods.

cs.CL↗
arXiv2023

Sufficient Control of Complex Networks

In this paper, we propose to study on sufficient control of complex networks which is to control a sufficiently large portion of the network, where only the quantity of controllable nodes matters. To the best of our knowledge, this is the first time that such a problem is investigated. We prove that the sufficient controllability problem can be converted into a minimum cost flow problem, for which an algorithm can be easily devised with polynomial complexity. Further, we study the problem of minimum-cost sufficient control, which is to drive a sufficiently large subset of the network nodes to any predefined state with the minimum cost using a given number of controllers. It is proved that the problem is NP-hard. We propose an ``extended $L_{\mathrm{0}}$-norm-constraint-based Projected Gradient Method" (eLPGM) algorithm which may achieve suboptimal solutions for the problems at small or medium sizes. To tackle the large-scale problems, we propose to convert the control problem into a graph algorithm problem, and devise an efficient low-complexity ``Evenly Divided Control Paths" (EDCP) algorithm to tackle the graph problem. Simulation results on both synthetic and real-life networks are provided, demonstrating the satisfactory performance of the proposed methods.

eess.SY↗
arXiv2023

Sharp constants for a class of linear integral operators on higher-dimensional product spaces

In this paper, we will study a class of linear integral operators with the nonnegative kernels on higher-dimensional product spaces, the norms of the operators can be obtained by integral of the product of the kernel function and finitely many basic functions. As application, we obtain the sharp constant for linear Hilbert operator on Higher-dimensional product spaces.

math.FA↗
arXiv2023

SPP-CNN: An Efficient Framework for Network Robustness Prediction

This paper addresses the robustness of a network to sustain its connectivity and controllability against malicious attacks. This kind of network robustness is typically measured by the time-consuming attack simulation, which returns a sequence of values that record the remaining connectivity and controllability after a sequence of node- or edge-removal attacks. For improvement, this paper develops an efficient framework for network robustness prediction, the spatial pyramid pooling convolutional neural network (SPP-CNN). The new framework installs a spatial pyramid pooling layer between the convolutional and fully-connected layers, overcoming the common mismatch issue in the CNN-based prediction approaches and extending its generalizability. Extensive experiments are carried out by comparing SPP-CNN with three state-of-the-art robustness predictors, namely a CNN-based and two graph neural networks-based frameworks. Synthetic and real-world networks, both directed and undirected, are investigated. Experimental results demonstrate that the proposed SPP-CNN achieves better prediction performances and better generalizability to unknown datasets, with significantly lower time-consumption, than its counterparts.

cs.LG↗
arXiv2023

FASTCloud: A novel framework of assessment and selection for trustworthy cloud service

By virtue of technology and benefit advantages, cloud computing has increasingly attracted a large number of potential cloud consumers (PCC) plan to migrate the traditional business to the cloud service. However, trust has become one of the most challenging issues that prevent the PCC from adopting cloud services, especially in trustworthy cloud service selection. Besides, due to the diversity and dynamic of quality of service (QoS) in the cloud environment, the existing trust assessment methods based on the single constant value of QoS attribute and the subjective weight assignment are not good enough to provide an effective solution for PCCs to identify and select a trustworthy cloud service among a wide range of functionally-equivalent cloud service providers (CSPs). To address the challenge, a novel assessment and selection framework for trustworthy cloud service, FASTCloud, is proposed in this study. This framework facilitates PCCs to select a trustworthy cloud service based on their actual QoS requirements. In order to accurately and efficiently assess the trust level of cloud services, a QoS-based trust assessment model is proposed. This model represents a trust level assessment method based on the interval multiple attributes with an objective weight assignment method based on the deviation maximization to adaptively determine the trust level of different cloud services provisioned by candidate CSPs. The advantage of the proposed trust level assessment method in time complexity is demonstrated by the performance analysis and comparison. The experimental result of a case study with an open-source dataset shows that the trust model is efficient in cloud service trust assessment and the FASTCloud can effectively help PCCs select a trustworthy cloud service.

cs.DC↗
arXiv2023

Dual Intent Enhanced Graph Neural Network for Session-based New Item Recommendation

Recommender systems are essential to various fields, e.g., e-commerce, e-learning, and streaming media. At present, graph neural networks (GNNs) for session-based recommendations normally can only recommend items existing in users' historical sessions. As a result, these GNNs have difficulty recommending items that users have never interacted with (new items), which leads to a phenomenon of information cocoon. Therefore, it is necessary to recommend new items to users. As there is no interaction between new items and users, we cannot include new items when building session graphs for GNN session-based recommender systems. Thus, it is challenging to recommend new items for users when using GNN-based methods. We regard this challenge as '\textbf{G}NN \textbf{S}ession-based \textbf{N}ew \textbf{I}tem \textbf{R}ecommendation (GSNIR)'. To solve this problem, we propose a dual-intent enhanced graph neural network for it. Due to the fact that new items are not tied to historical sessions, the users' intent is difficult to predict. We design a dual-intent network to learn user intent from an attention mechanism and the distribution of historical data respectively, which can simulate users' decision-making process in interacting with a new item. To solve the challenge that new items cannot be learned by GNNs, inspired by zero-shot learning (ZSL), we infer the new item representation in GNN space by using their attributes. By outputting new item probabilities, which contain recommendation scores of the corresponding items, the new items with higher scores are recommended to users. Experiments on two representative real-world datasets show the superiority of our proposed method. The case study from the real-world verifies interpretability benefits brought by the dual-intent module and the new item reasoning module. The code is available at Github: https://github.com/Ee1s/NirGNN

cs.IR↗
arXiv2023

Joint Task Offloading and Resource Allocation for Streaming Application in Cooperative Mobile Edge Computing

Mobile edge computing (MEC) enables resource-limited IoT devices to complete computation-intensive or delay-sensitive task by offloading the task to adjacent edge server deployed at the base station (BS), thus becoming an important technology in 5G and beyond. Due to channel occlusion, some users may not be able to access the computation capability directly from the BS. Confronted with this issue, many other devices in the MEC system can serve as cooperative nodes to collect the tasks of these users and further forward them to the BS. In this paper, we study a MEC system in which multiple users continuously generate the tasks and offload the tasks to the BS through a cooperative node. As the tasks are continuously generated, users should simultaneously execute the task generation in the current time frame and the task offloading of the last time frame, i.e. the task is processed in a streaming model. To optimize the power consumption of the users and the cooperative node for finishing these streaming tasks, we investigate the duration of each step in finishing the tasks together with multiuser offloading ratio and bandwidth allocation within two cases: the BS has abundant computation capacity (Case I) and the BS has limited computation capacity (Case II). For both cases, the formulated optimization problems are nonconvex due to fractional structure of the objective function and complicated variable coupling. To address this issue, we propose optimal solution algorithm with low complexity. Finally, simulation is carried out to verify the effectiveness of the proposed methods and reveal the performance of the considered system.

cs.IT↗