检索结果-内蒙古大学图书馆

Controllable Financial Market Generation with Diffusion Guided Meta Agent

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Huang, Yu-Hao Xu, Chang Liu, Yang Liu, Weiqing Li, Wu-Jun Bian, Jiang National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University China Microsoft Research Asia China

Order flow modeling stands as the most fundamental and essential financial task, as orders embody the minimal unit within a financial market. However, current approaches often result in unsatisfactory fidelity in generating order flow, and their generation lacks controllability, thereby limiting their application scenario. In this paper, we advocate incorporating controllability into the market generation process, and propose a Diffusion Guided meta Agent (DiGA) model to address the problem. Specifically, we utilize a diffusion model to capture dynamics of market state represented by time-evolving distribution parameters about mid-price return rate and order arrival rate, and define a meta agent with financial economic priors to generate orders from the corresponding distributions. Extensive experimental results demonstrate that our method exhibits outstanding controllability and fidelity in generation. Furthermore, we validate DiGA’s effectiveness as generative environment for downstream financial applications. © 2024, CC BY.

关键词： Commerce

On the Optimal Batch Size for Byzantine-Robust Distributed Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yang, Yi-Rui Shi, Chang-Wei Li, Wu-Jun National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University China

Byzantine-robust distributed learning (BRDL), in which computing devices are likely to behave abnormally due to accidental failures or malicious attacks, has recently become a hot research topic. However, even in the independent and identically distributed (i.i.d.) case, existing BRDL methods will suffer from a significant drop on model accuracy due to the large variance of stochastic gradients. Increasing batch sizes is a simple yet effective way to reduce the variance. However, when the total number of gradient computation is fixed, a too-large batch size will lead to a too-small iteration number (update number), which may also degrade the model accuracy. In view of this challenge, we mainly study the optimal batch size when the total number of gradient computation is fixed in this work. In particular, we theoretically and empirically show that when the total number of gradient computation is fixed, the optimal batch size in BRDL increases with the fraction of Byzantine workers. Therefore, compared to the case without attacks, the batch size should be set larger when under Byzantine attacks. However, for existing BRDL methods, large batch sizes will lead to a drop on model accuracy, even if there is no Byzantine attack. To deal with this problem, we propose a novel BRDL method, called Byzantine-robust stochastic gradient descent with normalized momentum (ByzSGDnm), which can alleviate the drop on model accuracy in large-batch cases. Moreover, we theoretically prove the convergence of ByzSGDnm for general non-convex cases under Byzantine attacks. Empirical results show that ByzSGDnm has a comparable performance to existing BRDL methods under bit-flipping failure, but can outperform existing BRDL methods under deliberately crafted attacks. Copyright © 2023, The Authors. All rights reserved.

关键词： Drops

FedREP: A Byzantine-Robust, Communication-Efficient and Privacy-Preserving Framework for Federated Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yang, Yi-Rui Wang, Kun Li, Wu-Jun National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University China

Federated learning (FL) has recently become a hot research topic, in which Byzantine robustness, communication efficiency and privacy preservation are three important aspects. However, the tension among these three aspects makes it hard to simultaneously take all of them into account. In view of this challenge, we theoretically analyze the conditions that a communication compression method should satisfy to be compatible with existing Byzantine-robust methods and privacy-preserving methods. Motivated by the analysis results, we propose a novel communication compression method called consensus sparsification (ConSpar). To the best of our knowledge, ConSpar is the first communication compression method that is designed to be compatible with both Byzantine-robust methods and privacy-preserving methods. Based on ConSpar, we further propose a novel FL framework called FedREP, which is Byzantine-robust, communication-efficient and privacy-preserving. We theoretically prove the Byzantine robustness and the convergence of FedREP. Empirical results show that FedREP can significantly outperform communication-efficient privacy-preserving baselines. Furthermore, compared with Byzantine-robust communication-efficient baselines, FedREP can achieve comparable accuracy with the extra advantage of privacy preservation. Copyright © 2023, The Authors. All rights reserved.

关键词： Privacy-preserving techniques

SOFTENED SYMBOL GROUNDING FOR NEURO-SYMBOLIC SYSTEMS

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Zenan Yao, Yuan Chen, Taolue Xu, Jingwei Cao, Chun Ma, Xiaoxing Lü, Jian State Key Lab of Novel Software Technology Department of Computer Science and Technology Nanjing University China Department of Computer Science Birkbeck University of London United Kingdom

Neuro-symbolic learning generally consists of two separated worlds, i.e., neural network training and symbolic constraint solving, whose success hinges on symbol grounding, a fundamental problem in AI. This paper presents a novel, softened symbol grounding process, bridging the gap between the two worlds, and resulting in an effective and efficient neuro-symbolic learning framework. Technically, the framework features (1) modeling of symbol solution states as a Boltzmann distribution, which avoids expensive state searching and facilitates mutually beneficial interactions between network training and symbolic reasoning;(2) a new MCMC technique leveraging projection and SMT solvers, which efficiently samples from disconnected symbol solution spaces;(3) an annealing mechanism that can escape from sub-optimal symbol groundings. Experiments with three representative neuro-symbolic learning tasks demonstrate that, owining to its superior symbol grounding capability, our framework successfully solves problems well beyond the frontier of the existing proposals. © 2024, CC BY.

关键词： Boltzmann equation

Actor of Things: Resilient and Efficient Distributed AIoT Applications in an Actor System

学校读者我要写书评

暂无评论

Actor of Things: Resilient and Efficient Distributed AIoT Ap...

Asia-Pacific Conference on software Engineering

作者： Tianqi Ren Chun Cao Jun Ma Department of Computer Science and Technology State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

ISBN: (数字)9798331534011

ISBN: (纸本)9798331534028

Complex and distributed AIoT applications usually involve multiple interconnected heterogeneous devices, leading to requirements of modeling device states and interaction diagrams. It is found that actor model is well-suited for distributed IoT contexts due to its inherent distributive nature and concurrency safety, but existing works lack runtime scalability and support for asynchronous task execution. In this paper, we propose Actor of Things, an approach to transform AIoT applications into efficient and resilient actor systems. We first transform synchronous function calls into Actor Task Graph which decouples nodes with real actors and executes asynchronously. The graph is then executed by Actor Runtime which deploys and schedules actors on IoT devices committed to lowest latency. Preliminary evaluations of our implementation demonstrate the advantages of proposed method in building distributed AIoT applications.

关键词： Concurrent computing Schedules Runtime software architecture Scalability Buildings Transforms Safety Internet of Things Context modeling

Adaptively feature matching via joint transformational-spatial clustering

学校读者我要写书评

暂无评论

Adaptively feature matching via joint transformational-spati...

作者： Wang, Linbo Tan, Li Fang, Xianyong Guo, Yanwen Wan, Shaohua MOE Key Laboratory of Intelligent Computing and Signal Processing School of Computer Science and Technology Anhui University Hefei China National Key Lab for Novel Software Technology Nanjing University Nanjing China School of Information and Safety Engineering Zhongnan University of Economics and Law Wuhan China

The transformational and spatial proximities are important cues for identifying inliers from an appearance based match set because correct matches generally stay close in input images and share similar local transformations. However, most existing approaches only check one type of them or both types consecutively with manually set thresholds, and thus their matching accuracy and flexibility in handling large-scale images are limited. In this paper, we present an efficient clustering based approach to identify match inliers with both proximities simultaneously. It first projects the putative matches into a joint transformational-spatial space, where mismatches tend to scatter all around while correct matches gather together. A mode-seeking process based on joint kernel density estimation is then proposed to obtain significant clusters in the joint space, where each cluster contains matches mapping the same object across images with high accuracy. Moreover, kernel bandwidths for measuring match proximities are adaptively set during density estimation, which enhances its applicability for matching different images. Experiments on three standard datasets show that the proposed approach delivers superior performance on a variety of feature matching tasks, including multi-object matching, duplicate object matching and object retrieval. © 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Clustering Density estimation Feature matching Mode-seeking

RAT: A Refactoring-Aware Tool for Tracking Code History 24

学校读者我要写书评

暂无评论

RAT: A Refactoring-Aware Tool for Tracking Code History

46th International Conference on software Engineering: Companion, ICSE-Companion 2024

作者： Niu, Feifei Shao, Junqian Xu, Chaofan Mayr-Dorn, Christoph Assunção, Wesley K. G. Huang, Liguo Li, Chuanyi Ge, Jidong Luo, Bin Egyed, Alexander State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Institute for Software Systems Engineering Johannes Kepler University Linz Austria Department of Computer Science North Carolina State University North Carolina United States Department of Computer Science and Engineering Southern Methodist University DallasTX United States

ISBN: (纸本)9798400705021

History of code elements is essential for software maintenance tasks. However, code refactoring is one of the main causes that makes obtaining a consistent view on code evolution difficult as renaming or moving source code elements break such history. To this end, this paper presents RAT, a refactoring-aware tool for keeping track of code elements evolution across time, not just in terms of revisions but also in terms of refactoring. This is the first tool that enables fine-grained code element traceability of the whole repository. Empirical evaluation of leveraging our tool in three bug localization techniques relying on code history shows significant improvement in localization accuracy. Based on our findings, we believe that many of the state-of-the-art approaches using past source code data would benefit from our tool. Demo Tool: https://***/feifeiniu-se/RAT-Demo. Demo Video: https://***/VI-xwUaIPp4. © 2024 IEEE computer Society. All rights reserved.

关键词： Rats

MACRO: Incentivizing Multi-Leader Game-Based Pareto-Efficient Crowdsourcing for Video Analytics 40

学校读者我要写书评

暂无评论

MACRO: Incentivizing Multi-Leader Game-Based Pareto-Efficien...

40th IEEE International Conference on Data Engineering, ICDE 2024

作者： Chen, Yu Zhang, Sheng Zhou, Ziying Wang, Xiaokun Liang, Yu Chen, Ning Yan, Yuting Xiao, Mingjun Wu, Jie Qian, Zhuzhong Xu, Harry Nanjing University State Key Lab. for Novel Software Technology China School of Systems Science Beijing Normal University China School of Computer and Electronic Information School of Artificial Intelligence Nanjing Normal University China School of Computer Science and Technology University of Science and Technology of China China Temple University Center for Networked Computing United States Samueli School of Engineering University of California Computer Science Department Los Angeles United States

ISBN: (纸本)9798350317152

In recent years, many crowdsourcing platforms have emerged, using the resources of recruited workers to perform diverse outsourcing tasks, where the video analytics attracts much attention due to its practical implications. For maximum profits, platforms carefully choose the workers and determine the video analytics configurations to ensure accuracy;meanwhile, workers possess the flexibility to tailor the configurations for their indivi-dual gains, which makes it hard for platforms to optimize their profits considering the platform-worker conflicts. In this paper, we design an incentive mechanism for Multi-leader game-based video Analytics upon CROwdsourcing, named MACRO, to over-come the above situation. Under that mechanism, we first formu-late the utility optimization problems for platforms and workers, respectively. We then propose a dual ascent-based method to op-timally determine the video analytics configurations for a multi-platform game, ensuring Pareto efficiency. Moreover, in the context of a multi-leader game involving platform-worker conflicts, we design an incentive function with its incentive factor update strategy and propose an ADMM-based approach for maximizing incentives that motivate workers to contribute to the platforms' profits. Rigorous proofs demonstrate the linear convergence of the MACRO to the multi-leader Stackelberg equilibrium. Trace-driven experiments show that MACRO improves the Pareto efficiency by 26.3%, outperforming other approaches. © 2024 IEEE.

关键词： Profitability

Graph Contrastive Learning with Cohesive Subgraph Awareness 24

学校读者我要写书评

暂无评论

Graph Contrastive Learning with Cohesive Subgraph Awareness

33rd ACM Web Conference, WWW 2024

作者： Wu, Yucheng Wang, Leye Han, Xiao Ye, Han-Jia Key Lab of High Confidence Software Technologies Peking University Ministry of Education School of Computer Science Peking University Beijing China School of Information Management and Engineering Shanghai University of Finance and Economics Shanghai China National Key Laboratory for Novel Software Technology Nanjing University School of Artificial Intelligence Nanjing University Nanjing China

ISBN: (纸本)9798400701719

Graph contrastive learning (GCL) has emerged as a state-of-the-art strategy for learning representations of diverse graphs including social and biomedical networks. GCL widely uses stochastic graph topology augmentation, such as uniform node dropping, to generate augmented graphs. However, such stochastic augmentations may severely damage the intrinsic properties of a graph and deteriorate the following representation learning process. We argue that incorporating an awareness of cohesive subgraphs during the graph augmentation and learning processes has the potential to enhance GCL performance. To this end, we propose a novel unified framework called CTAug, to seamlessly integrate cohesion awareness into various existing GCL mechanisms. In particular, CTAug comprises two specialized modules: topology augmentation enhancement and graph learning enhancement. The former module generates augmented graphs that carefully preserve cohesion properties, while the latter module bolsters the graph encoder's ability to discern subgraph patterns. Theoretical analysis shows that CTAug can strictly improve existing GCL mechanisms. Empirical experiments verify that CTAug can achieve state-of-the-art performance for graph representation learning, especially for graphs with high degrees. The code is available at https://***/10.5281/zenodo.10594093, or https://***/wuyucheng2002/CTAug. © 2024 ACM.

关键词： Stochastic systems