检索结果-内蒙古大学图书馆

OPTIMAL SETTINGS FOR CRYPTOCURRENCY TRADING PaiRS

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhang, Di Zhou, Youzhou School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou215123 China School of Mathematics and Physics Xi’an Jiaotong-Liverpool University Suzhou215123 China

The goal of cryptocurrencies is decentralization, but it is impractical to set up a trading market between every two currencies. To solve this optimization problem, we use a two-stage process: 1) Fill in missing values based on a regularized, truncated eigenvalue decomposition, where the regularization term is used to control what extent missing values should be limited to zero. 2) Search for the optimal trading pairs, based on a branch and bound process, with heuristic search and pruning strategies. The experimental results show that: 1) If the number of denominated coins is not limited, we will get a more decentralized trading pair settings, which advocates the establishment of trading pairs directly between large currency pairs. 2) There is a certain room for optimization in all exchanges. The setting of inappropriate trading pairs is mainly caused by subjectively setting small coins to quote, or failing to track emerging big coins in time. 3) Too few trading pairs will lead to low coverage;too many trading pairs will need to be adjusted with markets frequently. Exchanges should consider striking an appropriate balance between them. Copyright © 2022, The Authors. All rights reserved.

关键词： Heuristic algorithms

Combining Transformer based Deep Reinforcement Learning with Black-Litterman Model for Portfolio Optimization

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Sun, Ruoyu Stefanidis, Angelos Jiang, Zhengyong Su, Jionglong Xi’an Jiaotong-Liverpool University School of Mathematics and Physics Department of Financial and Actuarial Mathematics Suzhou215123 China School of AI and Advanced Computing Suzhou215412 China

As a model-free algorithm, deep reinforcement learning (DRL) agent learns and makes decisions by interacting with the environment in an unsupervised way. In recent years, DRL algorithms have been widely applied by scholars for portfolio optimization in consecutive trading periods, since the DRL agent can dynamically adapt to market changes and does not rely on the specification of the joint dynamics across the assets. However, typical DRL agents for portfolio optimization cannot learn a policy that is aware of the dynamic correlation between portfolio asset returns. Since the dynamic correlations among portfolio assets are crucial in optimizing the portfolio, the lack of such knowledge makes it difficult for the DRL agent to maximize the return per unit of risk, especially when the target market permits short selling (i.e., the US stock market). In this research, we propose a hybrid portfolio optimization model combining the DRL agent and the Black-Litterman (BL) model to enable the DRL agent to learn the dynamic correlation between the portfolio asset returns and implement an efficacious long/short strategy based on the correlation. Essentially, the DRL agent is trained to learn the policy to apply the BL model to determine the target portfolio weights. In this model, we formulate a specific objective function based on the environment’s reward function, which considers the return, risk, and transaction scale of the portfolio. Our DRL agent is trained by propagating the objective function’s gradient to the policy function of our DRL agent. To test our DRL agent, we construct the portfolio based on all the Dow Jones Industrial Average constitute stocks. Empirical results of the experiments conducted on real-world United States stock market data demonstrate that our DRL agent significantly outperforms various comparison portfolio choice strategies and alternative DRL frameworks by at least 42% in terms of accumulated return. In terms of the return per unit of risk, ou

关键词： Reinforcement learning

MODELING RANDOMLY WALKING VOLATILITY WITH CHaiNED GAMMA DISTRIBUTIONS

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhang, Di Niu, Qiang Zhou, Youzhou School of AI and Advanced Computing Xi'an Jiaotong-Liverpool University Suzhou215123 China School of Mathematics and Physics Xi'an Jiaotong-Liverpool University Suzhou215123 China

Volatility clustering is a common phenomenon in financial time series. Typically, linear models can be used to describe the temporal autocorrelation of the (logarithmic) variance of returns. Considering the difficulty in estimating this model, we construct a Dynamic Bayesian Network, which utilizes the conjugate prior relation of normal-gamma and gamma-gamma, so that its posterior form locally remains unchanged at each node. This makes it possible to find approximate solutions using variational methods quickly. Furthermore, we ensure that the volatility expressed by the model is an independent incremental process after inserting dummy gamma nodes between adjacent time steps. We have found that this model has two advantages: 1) It can be proved that it can express heavier tails than Gaussians, i.e., have positive excess kurtosis, compared to popular linear models. 2) If the variational inference(VI) is used for state estimation, it runs much faster than Monte Carlo(MC) methods since the calculation of the posterior uses only basic arithmetic operations. And its convergence process is deterministic. We tested the model, named Gam-Chain, using recent Crypto, Nasdaq, and Forex records of varying resolutions. The results show that: 1) In the same case of using MC, this model can achieve comparable state estimation results with the regular lognormal chain. 2) In the case of only using VI, this model can obtain accuracy that are slightly worse than MC, but still acceptable in practice;3) Only using VI, the running time of Gam-Chain, in general case, can be reduced to below 5% of that based on the lognormal chain via *** Codes 91-10 Copyright © 2022, The Authors. All rights reserved.

关键词： State estimation

MDAEN: Multi-Dimensional Attention-based Ensemble Network in Deep Reinforcement Learning Framework for Portfolio Management

学校读者我要写书评

暂无评论

MDAEN: Multi-Dimensional Attention-based Ensemble Network in...

International Conference on Cyber-Enabled Distributed computing and Knowledge Discovery, CyberC

作者： Ruiyu Zhang Xiaotian Ren Fengchen Gu Angelos Stefanidis Ruoyu Sun Jionglong Su School of AI and Advanced Computing XJTLU Entrepreneur College (Taicang) School of Mathematics and Physics Xi’an Jiaotong-Liverpool University Xi’an Jiaotong-Liverpool University Suzhou China

Reinforcement Learning algorithms are widely applied in many diverse fields, including portfolio management. Ensemble of Identical Independent Evaluators (EIIE) framework proposed by Jiang et al. achieved portfolio management based on their deep reinforcement learning algorithm. In the implementation of EIIE framework, a neural network such as the Convolutional Neural Network is applied as the policy network, to uncover more patterns in the data. However, this network typology is inefficient due to its simple structure. To overcome the shortcoming of EIIE framework, this paper introduces a novel algorithm, the Multi-Dimensional Attention-based Ensemble Network (MDAEN) strategy, which consists of a features-attention module and an assets-attention module. The MDAEN applies different types of attention mechanisms to extract information from the assets. Having adopted the reinforcement learning framework from Jiang et al., the agent is able to process transactions through MDAEN in a market. In our portfolio establishment, Bitcoin together with eleven other cryptocurrencies is selected to validate the performance of MDAEN against seven traditional portfolio strategies and EIIE. The experimental result demonstrates the efficacy of our strategy outperforming all other strategies by at least 35% in profitability and at least 30% in Sharpe Ratio.

关键词： Deep learning Profitability Neural networks Reinforcement learning Knowledge discovery Feature extraction Data mining

Bio-inspired multi-layer spiking neural network extracts discriminative features from speech signals

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Tavanaei, Amirhossein Maida, Anthony Center for Advanced Computer Studies Bio-inspired AI Lab School of Computing and Informatics University of Louisiana at Lafayette LafayetteLA70504 United States

Spiking neural networks (SNNs) enable power-efficient implementations due to their sparse, spike-based coding scheme. This paper develops a bio-inspired SNN that uses unsupervised learning to extract discriminative features from speech signals, which can subsequently be used in a classifier. The architecture consists of a spiking convolutional/pooling layer followed by a fully connected spiking layer for feature discovery. The convolutional layer of leaky, integrate-and-fire (LIF) neurons represents primary acoustic features. The fully connected layer is equipped with a probabilistic spike-timing-dependent plasticity learning rule. This layer represents the discriminative features through probabilistic, LIF neurons. To assess the discriminative power of the learned features, they are used in a hidden Markov model (HMM) for spoken digit recognition. The experimental results show performance above 96% that compares favorably with popular statistical feature extraction methods. Our results provide a novel demonstration of unsupervised feature acquisition in an SNN. Copyright © 2017, The Authors. All rights reserved.

关键词： Feature extraction

Generalized-Extended-State-Observer and Equivalent-Input-Disturbance Methods for Active Disturbance Rejection: Deep Observation and Comparison

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2023年第4期10卷 957-968页

作者： Jinhua She Kou Miyamoto Qing-Long Han Min Wu Hiroshi Hashimoto Qing-Guo Wang School of Engineering Tokyo University of TechnologyHachiojiTokyo 192-0982Japan K.Miyamoto is with the Institute of Technology Shimizu CorporationKotoTokyo 135-0044Japan School of Science Computing and Engineering TechnologiesSwinburne University of TechnologyMelbourneVIC 3122Australia School of Automation China University of GeosciencesWuhan 430074 Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of EducationWuhan 430074China School of Industrial Technology Advanced Institute of Industrial TechnologyTokyo 140-0011Japan Institute of Artificial Intelligence and Future Networks Beijing Normal UniversityZhuhai 519087 Guangdong Key Lab of AI and Multi-Modal Data Processing Guangdong Provincial Key Laboratory of Interdisciplinary Research and Application for Data Science BNUHKBU United International College Zhuhai 519087China

Active disturbance-rejection methods are effective in estimating and rejecting disturbances in both transient and steady-state *** paper presents a deep observation on and a comparison between two of those methods:the generalized extended-state observer(GESO)and the equivalent input disturbance(EID)from assumptions,system configurations,stability conditions,system design,disturbance-rejection performance,and extensibility.A time-domain index is introduced to assess the disturbance-rejection performance.A detailed observation of disturbance-suppression mechanisms reveals the superiority of the EID approach over the GESO method.A comparison between these two methods shows that assumptions on disturbances are more practical and the adjustment of disturbance-rejection performance is easier for the EID approach than for the GESO method.

关键词： Active disturbance-rejection control(ADRC) disturbance observer(DOB) equivalent input disturbance(EID) extendedstate observer(ESO) generalized extended-state observer(GESO)

Attention-Based Multimodal Bilinear Feature Fusion for Lung Cancer Survival Analysis

学校读者我要写书评

暂无评论

Attention-Based Multimodal Bilinear Feature Fusion for Lung ...

IEEE Symposium on Bioinformatics and Bioengineering (BIBE)

作者： Hongbin Na Lilin Wang Xinyao Zhuang Jianfei He Zhenyu Liu Zimu Wang Hong-Seng Gan School of Computer Science and Engineering University of New South Wales Sydney Australia School of AI and Advanced Computing Xi'an Jiaotong-Liverpool University Suzhou China School of Advanced Technology Xi'an Jiaotong-Liverpool University Suzhou China

Survival analysis (SA) is an essential task that aims to predict survival status and duration, determine individual and precise treatment strategies, and assess disease intensity and direction. However, the current research on multimodal SA has identified three unique challenges: inefficient cross-modal information integration, insufficient inter-modal key features, and noisy data. In this paper, we propose a novel SA framework, named Attention-based Multimodal Bilinear Feature Fusion (AMBF)-SA, to address the aforementioned challenges. Specifically, AMBF-SA first performs feature extraction with the off-the-shelf models on each modality separately, then fuses the features between multiple sources and modalities using our proposed AMBF method, and finally outputs the survival prediction by a multi-layer perception (MLP). Experimental results on the Non-small Cell Lung Cancer (NSCLC) Radiogenomics dataset demonstrate remark performance of AMBF-SA compared with the rest of the experimented models, including the models trained with single and combined modalities under the Mean Absolute Error (MAE) and the Concordance Index (C-index) evaluation metrics, indicating the usefulness of our proposed framework.

关键词：

Generating Valid and Natural Adversarial Examples with Large Language Models

学校读者我要写书评

暂无评论

Generating Valid and Natural Adversarial Examples with Large...

International Conference on Computer Supported Cooperative Work in Design

作者： Zimu Wang Wei Wang Qi Chen Qiufeng Wang Anh Nguyen School of Advanced Technology Xi’an Jiaotong-Liverpool University Suzhou China Department of Computer Science University of Liverpool Liverpool United Kingdom School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Deep learning-based natural language processing (NLP) models, particularly pre-trained language models (PLMs), have been revealed to be vulnerable to adversarial attacks. However, the adversarial examples generated by many mainstream word-level adversarial attack models are neither valid nor natural, leading to the loss of semantic maintenance, grammaticality, and human imperceptibility. Based on the exceptional capacity of language understanding and generation of large language models (LLMs), we propose LLM-Attack, which aims at generating both valid and natural adversarial examples with LLMs. The method consists of two stages: word importance ranking (which searches for the most vulnerable words) and word synonym replacement (which substitutes them with their synonyms obtained from LLMs). Experimental results on the Movie Review (MR), IMDB, and Yelp Review Polarity datasets against the baseline adversarial attack models illustrate the effectiveness of LLM-Attack, and it outperforms the baselines in human and GPT-4 evaluation by a significant margin. The model can generate adversarial examples that are typically valid and natural, with the preservation of semantic meaning, grammaticality, and human imperceptibility.

关键词： Reviews Federated learning Large language models Computational modeling Semantics Motion pictures Natural language processing

Application of Voformer-EC Clustering Algorithm to Stock Multivariate Time Series Data

学校读者我要写书评

暂无评论

Application of Voformer-EC Clustering Algorithm to Stock Mul...

International Conference on Cyber-Enabled Distributed computing and Knowledge Discovery, CyberC

作者： Ning Xin Shaheen Khatoon Md Maruf Hasan School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou China School of Arch. Comp. and Engineering University of East London London United Kingdom

Clustering stocks based on similar increasing and decreasing trends pose a challenging problem in stock forecasting. Despite the extensive research on stock forecasting, striking a balance between effective clustering and computational speed remains an ongoing challenge. Traditional multivariate time series clustering methods are difficult to guarantee high speed with high accuracy. This study introduces the Voformer-EC model as a novel approach to address this issue, enhancing the analysis of multivariate time series data related to stocks. The Voformer-EC model incorporates time features and volatility, leveraging the Voformer neural network to extract time-related features and perform clustering. To evaluate its effectiveness, we applied the model to Nifty 50 Index data recorded every 60 minutes from February 2nd to February 28th, 2015, and compared it with a traditional approach. The results demonstrate a significant improvement in clustering accuracy using the Voformer-EC model. Building on these promising outcomes, future research aims to explore the application of the Voformer-EC model to temperature and precipitation data for identifying drought-prone areas. This implementation will enable targeted risk mitigation strategies to be employed effectively, advancing precision in addressing climate-related challenges.

关键词：