Searcharxiv⌕ Search

arXiv subjects

Li Su

Publications and source records attributed to Li Su.

At least 55 records · Page 3Linked to original sources
arXiv2023

GraphScope Flex: LEGO-like Graph Computing Stack

Graph computing has become increasingly crucial in processing large-scale graph data, with numerous systems developed for this purpose. Two years ago, we introduced GraphScope as a system addressing a wide array of graph computing needs, including graph traversal, analytics, and learning in one system. Since its inception, GraphScope has achieved significant technological advancements and gained widespread adoption across various industries. However, one key lesson from this journey has been understanding the limitations of a "one-size-fits-all" approach, especially when dealing with the diversity of programming interfaces, applications, and data storage formats in graph computing. In response to these challenges, we present GraphScope Flex, the next iteration of GraphScope. GraphScope Flex is designed to be both resource-efficient and cost-effective, while also providing flexibility and user-friendliness through its LEGO-like modularity. This paper explores the architectural innovations and fundamental design principles of GraphScope Flex, all of which are direct outcomes of the lessons learned during our ongoing development process. We validate the adaptability and efficiency of GraphScope Flex with extensive evaluations on synthetic and real-world datasets. The results show that GraphScope Flex achieves 2.4X throughput and up to 55.7X speedup over other systems on the LDBC Social Network and Graphalytics benchmarks, respectively. Furthermore, GraphScope Flex accomplishes up to a 2,400X performance gain in real-world applications, demonstrating its proficiency across a wide range of graph computing scenarios with increased effectiveness.

cs.DC↗
arXiv2023

Adapting pretrained speech model for Mandarin lyrics transcription and alignment

The tasks of automatic lyrics transcription and lyrics alignment have witnessed significant performance improvements in the past few years. However, most of the previous works only focus on English in which large-scale datasets are available. In this paper, we address lyrics transcription and alignment of polyphonic Mandarin pop music in a low-resource setting. To deal with the data scarcity issue, we adapt pretrained Whisper model and fine-tune it on a monophonic Mandarin singing dataset. With the use of data augmentation and source separation model, results show that the proposed method achieves a character error rate of less than 18% on a Mandarin polyphonic dataset for lyrics transcription, and a mean absolute error of 0.071 seconds for lyrics alignment. Our results demonstrate the potential of adapting a pretrained speech model for lyrics transcription and alignment in low-resource scenarios.

eess.AS↗
arXiv2023

Enhancing Motor Imagery Decoding in Brain Computer Interfaces using Riemann Tangent Space Mapping and Cross Frequency Coupling

Objective: Motor Imagery (MI) serves as a crucial experimental paradigm within the realm of Brain Computer Interfaces (BCIs), aiming to decoding motor intentions from electroencephalogram (EEG) signals. Method: Drawing inspiration from Riemannian geometry and Cross-Frequency Coupling (CFC), this paper introduces a novel approach termed Riemann Tangent Space Mapping using Dichotomous Filter Bank with Convolutional Neural Network (DFBRTS) to enhance the representation quality and decoding capability pertaining to MI features. DFBRTS first initiates the process by meticulously filtering EEG signals through a Dichotomous Filter Bank, structured in the fashion of a complete binary tree. Subsequently, it employs Riemann Tangent Space Mapping to extract salient EEG signal features within each sub-band. Finally, a lightweight convolutional neural network is employed for further feature extraction and classification, operating under the joint supervision of cross-entropy and center loss. To validate the efficacy, extensive experiments were conducted using DFBRTS on two well-established benchmark datasets: the BCI competition IV 2a (BCIC-IV-2a) dataset and the OpenBMI dataset. The performance of DFBRTS was benchmarked against several state-of-the-art MI decoding methods, alongside other Riemannian geometry-based MI decoding approaches. Results: DFBRTS significantly outperforms other MI decoding algorithms on both datasets, achieving a remarkable classification accuracy of 78.16% for four-class and 71.58% for two-class hold-out classification, as compared to the existing benchmarks.

q-bio.QM↗
arXiv2023

Popularity, face and voice: Predicting and interpreting livestreamers' retail performance using machine learning techniques

Livestreaming commerce, a hybrid of e-commerce and self-media, has expanded the broad spectrum of traditional sales performance determinants. To investigate the factors that contribute to the success of livestreaming commerce, we construct a longitudinal firm-level database with 19,175 observations, covering an entire livestreaming subsector. By comparing the forecasting accuracy of eight machine learning models, we identify a random forest model that provides the best prediction of gross merchandise volume (GMV). Furthermore, we utilize explainable artificial intelligence to open the black-box of machine learning model, discovering four new facts: 1) variables representing the popularity of livestreaming events are crucial features in predicting GMV. And voice attributes are more important than appearance; 2) popularity is a major determinant of sales for female hosts, while vocal aesthetics is more decisive for their male counterparts; 3) merits and drawbacks of the voice are not equally valued in the livestreaming market; 4) based on changes of comments, page views and likes, sales growth can be divided into three stages. Finally, we innovatively propose a 3D-SHAP diagram that demonstrates the relationship between predicting feature importance, target variable, and its predictors. This diagram identifies bottlenecks for both beginner and top livestreamers, providing insights into ways to optimize their sales performance.

econ.EM↗
arXiv2023

Self-supervised Cross-view Representation Reconstruction for Change Captioning

Change captioning aims to describe the difference between a pair of similar images. Its key challenge is how to learn a stable difference representation under pseudo changes caused by viewpoint change. In this paper, we address this by proposing a self-supervised cross-view representation reconstruction (SCORER) network. Concretely, we first design a multi-head token-wise matching to model relationships between cross-view features from similar/dissimilar images. Then, by maximizing cross-view contrastive alignment of two similar images, SCORER learns two view-invariant image representations in a self-supervised way. Based on these, we reconstruct the representations of unchanged objects by cross-attention, thus learning a stable difference representation for caption generation. Further, we devise a cross-modal backward reasoning to improve the quality of caption. This module reversely models a ``hallucination'' representation with the caption and ``before'' representation. By pushing it closer to the ``after'' representation, we enforce the caption to be informative about the difference in a self-supervised manner. Extensive experiments show our method achieves the state-of-the-art results on four datasets. The code is available at https://github.com/tuyunbin/SCORER.

cs.CV↗
arXiv2023

On a relationship between the characteristic and matching polynomials of a uniform hypertree

A hypertree is a connected hypergraph without cycles. Further a hypertree is called an $r$-tree if, additionally, it is $r$-uniform. Note that 2-trees are just ordinary trees. A classical result states that for any 2-tree $T$ with characteristic polynomial $ϕ_T(λ)$ and matching polynomial $φ_T(λ)$, then $ϕ_T(λ)=φ_T(λ).$ More generally, suppose $\mathcal{T}$ is an $r$-tree of size $m$ with $r\geq2$. In this paper, we extend the above classical relationship to $r$-trees and establish that \[ ϕ_{\mathcal{T}}(λ)=\prod_{H \sqsubseteq \mathcal{T}}φ_{H}(λ)^{a_{H}}, \] where the product is over all connected subgraphs $H$ of $\mathcal{T}$, and the exponent $a_{H}$ of the factor $φ_{H}(λ)$ can be written as \[ a_H=b^{m-e(H)-|\partial(H)|}c^{e(H)}(b-c)^{|\partial(H)|}, \] where $e(H)$ is the size of $H$, $\partial(H)$ is the boundary of $H$, and $b=(r-1)^{r-1}, c=r^{r-2}$. In particular, for $r=2$, the above correspondence reduces to the classical result for ordinary trees. In addition, we resolve a conjecture by Clark-Cooper [{\em Electron. J. Combin.}, 2018] and show that for any subgraph $H$ of an $r$-tree $\mathcal{T}$ with $r\geq3$, $φ_H(λ)$ divides $ϕ_{\mathcal{T}}(λ)$, and additionally $ϕ_H(λ)$ divides $ϕ_{\mathcal{T}}(λ)$, if either $r\geq 4$ or $H$ is connected when $r=3$. Moreover, a counterexample is given for the case when $H$ is a disconnected subgraph of a 3-tree.

math.CO↗
arXiv2023

Legion: Automatically Pushing the Envelope of Multi-GPU System for Billion-Scale GNN Training

Graph neural network(GNN) has been widely applied in real-world applications, such as product recommendation in e-commerce platforms and risk control in financial management systems. Several cache-based GNN systems have been built to accelerate GNN training in a single machine with multiple GPUs. However, these systems fail to train billion-scale graphs efficiently, which is a common challenge in the industry. In this work, we propose Legion, a system that automatically pushes the envelope of multi-GPU systems for accelerating billion-scale GNN training. First, we design a hierarchical graph partitioning mechanism that significantly improves the multi-GPU cache performance. Second, we build a unified multi-GPU cache that helps to minimize the PCIe traffic incurred by caching both graph topology and features with the highest hotness. Third, we develop an automatic caching management mechanism that adapts the multi-GPU cache plan according to the hardware specifications and various graphs to maximize the overall training throughput. Evaluations on various GNN models and multiple datasets show that Legion supports training billion-scale GNNs in a single machine and significantly outperforms the state-of-the-art cache-based systems on small graphs.

cs.DC↗
arXiv2023

A Phoneme-Informed Neural Network Model for Note-Level Singing Transcription

Note-level automatic music transcription is one of the most representative music information retrieval (MIR) tasks and has been studied for various instruments to understand music. However, due to the lack of high-quality labeled data, transcription of many instruments is still a challenging task. In particular, in the case of singing, it is difficult to find accurate notes due to its expressiveness in pitch, timbre, and dynamics. In this paper, we propose a method of finding note onsets of singing voice more accurately by leveraging the linguistic characteristics of singing, which are not seen in other instruments. The proposed model uses mel-scaled spectrogram and phonetic posteriorgram (PPG), a frame-wise likelihood of phoneme, as an input of the onset detection network while PPG is generated by the pre-trained network with singing and speech data. To verify how linguistic features affect onset detection, we compare the evaluation results through the dataset with different languages and divide onset types for detailed analysis. Our approach substantially improves the performance of singing transcription and therefore emphasizes the importance of linguistic features in singing analysis.

cs.SD↗
arXiv2023

Attacks Against Security Context in 5G Network

The security context used in 5G authentication is generated during the Authentication and Key Agreement (AKA) procedure and stored in both the user equipment (UE) and the network sides for the subsequent fast registration procedure. Given its importance, it is imperative to formally analyze the security mechanism of the security context. The security context in the UE can be stored in the Universal Subscriber Identity Module (USIM) card or in the baseband chip. In this work, we present a comprehensive and formal verification of the fast registration procedure based on the security context under the two scenarios in ProVerif. Our analysis identifies two vulnerabilities, including one that has not been reported before. Specifically, the security context stored in the USIM card can be read illegally, and the validity checking mechanism of the security context in the baseband chip can be bypassed. Moreover, these vulnerabilities also apply to 4G networks. As a consequence, an attacker can exploit these vulnerabilities to register to the network with the victim's identity and then launch other attacks, including one-tap authentication bypass leading to privacy disclosure, location spoofing, etc. To ensure that these attacks are indeed realizable in practice, we have responsibly confirmed them through experimentation in three operators. Our analysis reveals that these vulnerabilities stem from design flaws of the standard and unsafe practices by operators. We finally propose several potential countermeasures to prevent these attacks. We have reported our findings to the GSMA and received a coordinated vulnerability disclosure (CVD) number CVD-2022-0057.

cs.CR↗
arXiv2023

Neighborhood Contrastive Transformer for Change Captioning

Change captioning is to describe the semantic change between a pair of similar images in natural language. It is more challenging than general image captioning, because it requires capturing fine-grained change information while being immune to irrelevant viewpoint changes, and solving syntax ambiguity in change descriptions. In this paper, we propose a neighborhood contrastive transformer to improve the model's perceiving ability for various changes under different scenes and cognition ability for complex syntax structure. Concretely, we first design a neighboring feature aggregating to integrate neighboring context into each feature, which helps quickly locate the inconspicuous changes under the guidance of conspicuous referents. Then, we devise a common feature distilling to compare two images at neighborhood level and extract common properties from each image, so as to learn effective contrastive information between them. Finally, we introduce the explicit dependencies between words to calibrate the transformer decoder, which helps better understand complex syntax structure during training. Extensive experimental results demonstrate that the proposed method achieves the state-of-the-art performance on three public datasets with different change scenarios. The code is available at https://github.com/tuyunbin/NCT.

cs.CV↗
arXiv2022

ReAssigner: A Plug-and-Play Virtual Machine Scheduling Intensifier for Heterogeneous Requests

With the rapid development of cloud computing, virtual machine scheduling has become one of the most important but challenging issues for the cloud computing community, especially for practical heterogeneous request sequences. By analyzing the impact of request heterogeneity on some popular heuristic schedulers, it can be found that existing scheduling algorithms can not handle the request heterogeneity properly and efficiently. In this paper, a plug-and-play virtual machine scheduling intensifier, called Resource Assigner (ReAssigner), is proposed to enhance the scheduling efficiency of any given scheduler for heterogeneous requests. The key idea of ReAssigner is to pre-assign roles to physical resources and let resources of the same role form a virtual cluster to handle homogeneous requests. ReAssigner can cooperate with arbitrary schedulers by restricting their scheduling space to virtual clusters. With evaluations on the real dataset from Huawei Cloud, the proposed ReAssigner achieves significant scheduling performance improvement compared with some state-of-the-art scheduling methods.

cs.DC↗
arXiv2022

Preservation of the harmonic overtones in the violin family beyond Stradivari

Three hundred years ago, Antonio Stradivari enjoyed the golden period of the violin making in the human history. Luthiers and researchers endeavor to study his legendary legacy. Unfortunately, a consensus that the present progress has reached his level remains lacking. Most of the instruments suffer from low sounding power and enigmatic deficit of harmonic overtones. In fact, there is significant energy dissipation in the acoustic couplings in the violin family, completely unaware to luthiers and researchers. The current acoustic mechanism is inefficient in resonance of the wooden plates and hence is the low rate of mechanical energy conversion to the desired sound. We propose a new method for the amelioration of the acoustic couplings to enhance the preservation of the overtone components up to 161% of the pristine and to reduce the instrumental dissipation intensity up to 3.8 dB off (73% off). It results in the tremendously improved sound projection, the shorter reaction time, the longer reverberation time, and consequently the enrichment of the timbre complexity. Our physical method is applicable to the violin family of all sizes and ages, including the Stradivari, and extendable to all string instruments.

physics.class-ph↗
arXiv2022

Stability of Cournot duopoly games with isoelastic demands and quadratic costs

In this discussion draft, we explore different duopoly games of players with quadratic costs, where the market is supposed to have the isoelastic demand. Different from the usual approaches based on numerical computations, the methods used in the present work are built on symbolic computations, which can produce analytical and rigorous results. Our investigations show that the stability regions are enlarged for the games considered in this work compared to their counterparts with linear costs, which generalizes the classical results of "F. M. Fisher. The stability of the Cournot oligopoly solution: The effects of speeds of adjustment and increasing marginal costs. The Review of Economic Studies, 28(2):125--135, 1961.".

cs.SC↗
arXiv2022

Banyan: A Scoped Dataflow Engine for Graph Query Service

Graph query services (GQS) are widely used today to interactively answer graph traversal queries on large-scale graph data. Existing graph query engines focus largely on optimizing the latency of a single query. This ignores significant challenges posed by GQS, including fine-grained control and scheduling during query execution, as well as performance isolation and load balancing in various levels from across user to intra-query. To tackle these control and scheduling challenges, we propose a novel scoped dataflow for modeling graph traversal queries, which explicitly exposes concurrent execution and control of any subquery to the finest granularity. We implemented Banyan, an engine based on the scoped dataflow model for GQS. Banyan focuses on scaling up the performance on a single machine, and provides the ability to easily scale out. Extensive experiments on multiple benchmarks show that Banyan improves performance by up to three orders of magnitude over state-of-the-art graph query engines, while providing performance isolation and load balancing.

cs.DB↗
arXiv2021

Self-Regulated Learning for Egocentric Video Activity Anticipation

Future activity anticipation is a challenging problem in egocentric vision. As a standard future activity anticipation paradigm, recursive sequence prediction suffers from the accumulation of errors. To address this problem, we propose a simple and effective Self-Regulated Learning framework, which aims to regulate the intermediate representation consecutively to produce representation that (a) emphasizes the novel information in the frame of the current time-stamp in contrast to previously observed content, and (b) reflects its correlation with previously observed frames. The former is achieved by minimizing a contrastive loss, and the latter can be achieved by a dynamic reweighing mechanism to attend to informative frames in the observed content with a similarity comparison between feature of the current frame and observed frames. The learned final video representation can be further enhanced by multi-task learning which performs joint feature learning on the target activity labels and the automatically detected action and object class tokens. SRL sharply outperforms existing state-of-the-art in most cases on two egocentric video datasets and two third-person video datasets. Its effectiveness is also verified by the experimental fact that the action and object concepts that support the activity semantics can be accurately identified.

cs.CV↗
arXiv2021

Modeling Temporal Concept Receptive Field Dynamically for Untrimmed Video Analysis

Event analysis in untrimmed videos has attracted increasing attention due to the application of cutting-edge techniques such as CNN. As a well studied property for CNN-based models, the receptive field is a measurement for measuring the spatial range covered by a single feature response, which is crucial in improving the image categorization accuracy. In video domain, video event semantics are actually described by complex interaction among different concepts, while their behaviors vary drastically from one video to another, leading to the difficulty in concept-based analytics for accurate event categorization. To model the concept behavior, we study temporal concept receptive field of concept-based event representation, which encodes the temporal occurrence pattern of different mid-level concepts. Accordingly, we introduce temporal dynamic convolution (TDC) to give stronger flexibility to concept-based event analytics. TDC can adjust the temporal concept receptive field size dynamically according to different inputs. Notably, a set of coefficients are learned to fuse the results of multiple convolutions with different kernel widths that provide various temporal concept receptive field sizes. Different coefficients can generate appropriate and accurate temporal concept receptive field size according to input videos and highlight crucial concepts. Based on TDC, we propose the temporal dynamic concept modeling network (TDCMN) to learn an accurate and complete concept representation for efficient untrimmed video analysis. Experiment results on FCVID and ActivityNet show that TDCMN demonstrates adaptive event recognition ability conditioned on different inputs, and improve the event recognition performance of Concept-based methods by a large margin. Code is available at https://github.com/qzhb/TDCMN.

cs.CV↗
arXiv2021

Actions Speak Louder than Listening: Evaluating Music Style Transfer based on Editing Experience

The subjective evaluation of music generation techniques has been mostly done with questionnaire-based listening tests while ignoring the perspectives from music composition, arrangement, and soundtrack editing. In this paper, we propose an editing test to evaluate users' editing experience of music generation models in a systematic way. To do this, we design a new music style transfer model combining the non-chronological inference architecture, autoregressive models and the Transformer, which serves as an improvement from the baseline model on the same style transfer task. Then, we compare the performance of the two models with a conventional listening test and the proposed editing test, in which the quality of generated samples is assessed by the amount of effort (e.g., the number of required keyboard and mouse actions) spent by users to polish a music clip. Results on two target styles indicate that the improvement over the baseline model can be reflected by the editing test quantitatively. Also, the editing test provides profound insights which are not accessible from usual listening tests. The major contribution of this paper is the systematic presentation of the editing test and the corresponding insights, while the proposed music style transfer model based on state-of-the-art neural networks represents another contribution.

cs.SD↗
arXiv2021

ReconVAT: A Semi-Supervised Automatic Music Transcription Framework for Low-Resource Real-World Data

Most of the current supervised automatic music transcription (AMT) models lack the ability to generalize. This means that they have trouble transcribing real-world music recordings from diverse musical genres that are not presented in the labelled training data. In this paper, we propose a semi-supervised framework, ReconVAT, which solves this issue by leveraging the huge amount of available unlabelled music recordings. The proposed ReconVAT uses reconstruction loss and virtual adversarial training. When combined with existing U-net models for AMT, ReconVAT achieves competitive results on common benchmark datasets such as MAPS and MusicNet. For example, in the few-shot setting for the string part version of MusicNet, ReconVAT achieves F1-scores of 61.0% and 41.6% for the note-wise and note-with-offset-wise metrics respectively, which translates into an improvement of 22.2% and 62.5% compared to the supervised baseline model. Our proposed framework also demonstrates the potential of continual learning on new data, which could be useful in real-world applications whereby new data is constantly available.

cs.SD↗