Autonomous agents have long been a research focus in academic and industry communities. Previous research often focuses on training agents with limited knowledge within isolated environments, which diverges significantly from human learning processes, and makes the agents hard to achieve human-like decisions. Recently, through the acquisition of vast amounts of Web knowledge, large language models (LLMs) have shown potential in human-level intelligence, leading to a surge in research on LLM-based autonomous agents. In this paper, we present a comprehensive survey of these studies, delivering a systematic review of LLM-based autonomous agents from a holistic perspective. We first discuss the construction of LLM-based autonomous agents, proposing a unified framework that encompasses much of previous work. Then, we present a overview of the diverse applications of LLM-based autonomous agents in social science, natural science, and engineering. Finally, we delve into the evaluation strategies commonly used for LLM-based autonomous agents. Based on the previous studies, we also present several challenges and future directions in this field.

Despite significant successes achieved in knowledge discovery, traditional machine learning methods may fail to obtain satisfactory performances when dealing with complex data, such as imbalanced, high-dimensional, noisy data, etc. The reason behind is that it is difficult for these methods to capture multiple characteristics and underlying structure of data. In this context, it becomes an important topic in the data mining field that how to effectively construct an efficient knowledge discovery and mining model. Ensemble learning, as one research hot spot, aims to integrate data fusion, data modeling, and data mining into a unified framework. Specifically, ensemble learning firstly extracts a set of features with a variety of transformations. Based on these learned features, multiple learning algorithms are utilized to produce weak predictive results. Finally, ensemble learning fuses the informative knowledge from the above results obtained to achieve knowledge discovery and better predictive performance via voting schemes in an adaptive way. In this paper, we review the research progress of the mainstream approaches of ensemble learning and classify them based on different characteristics. In addition, we present challenges and possible research directions for each mainstream approach of ensemble learning, and we also give an extra introduction for the combination of ensemble learning with other machine learning hot spots such as deep learning, reinforcement learning, etc.

Having a formal model of neural networks can greatly help in understanding and verifying their properties, behavior, and response to external factors such as disease and medicine. In this paper, we adopt a formal model to represent neurons, some neuronal graphs, and their composition. Some specific neuronal graphs are known for having biologically relevant structures and behaviors and we call them archetypes. These archetypes are supposed to be the basis of typical instances of neuronal information processing. In this paper we study six fundamental archetypes (simple series, series with multiple outputs, parallel composition, negative loop, inhibition of a behavior, and contralateral inhibition), and we consider two ways to couple two archetypes: (i) connecting the output(s) of the first archetype to the input(s) of the second archetype and (ii) nesting the first archetype within the second one. We report and compare two key approaches to the formal modeling and verification of the proposed neuronal archetypes and some selected couplings. The first approach exploits the synchronous programming language Lustre to encode archetypes and their couplings, and to express properties concerning their dynamic behavior. These properties are verified thanks to the use of model checkers. The second approach relies on a theorem prover, the Coq Proof Assistant, to prove dynamic properties of neurons and archetypes.

Nowadays, smart buildings rely on Internet of things (IoT) technology derived from the cloud and fog computing paradigms to coordinate and collaborate between connected objects. Fog is characterized by low latency with a wider spread and geographically distributed nodes to support mobility, real-time interaction, and location-based services. To provide optimum quality of user life in modern buildings, we rely on a holistic Framework, designed in a way that decreases latency and improves energy saving and services efficiency with different capabilities. Discrete EVent system Specification (DEVS) is a formalism used to describe simulation models in a modular way. In this work, the sub-models of connected objects in the building are accurately and independently designed, and after installing them together, we easily get an integrated model which is subject to the fog computing Framework. Simulation results show that this new approach significantly, improves energy efficiency of buildings and reduces latency. Additionally, with DEVS, we can easily add or remove sub-models to or from the overall model, allowing us to continually improve our designs.

Finger vein biometrics have been extensively studied for the capability to detect aliveness, and the high security as intrinsic traits. However, vein pattern distortion caused by finger rotation degrades the performance of CNN in 2D finger vein recognition, especially in a contactless mode. To address the finger posture variation problem, we propose a 3D finger vein verification system extracting axial rotation invariant feature. An efficient 3D finger vein reconstruction optimization model is proposed and several accelerating strategies are adopted to achieve real-time 3D reconstruction on an embedded platform. The main contribution in this paper is that we are the first to propose a novel 3D point-cloud-based end-to-end neural network to extract deep axial rotation invariant feature, namely 3DFVSNet. In the network, the rotation problem is transformed to a permutation problem with the help of specially designed rotation groups. Finally, to validate the performance of the proposed network more rigorously and enrich the database resources for the finger vein recognition community, we built the largest publicly available 3D finger vein dataset with different degrees of finger rotation, namely the Large-scale Finger Multi-Biometric Database-3D Pose Varied Finger Vein (SCUT LFMB-3DPVFV) Dataset. Experimental results on 3D finger vein datasets show that our 3DFVSNet holds strong robustness against axial rotation compared to other approaches.

Hybrid memory systems composed of dynamic random access memory (DRAM) and Non-volatile memory (NVM) often exploit page migration technologies to fully take the advantages of different memory media. Most previous proposals usually migrate data at a granularity of 4 KB pages, and thus waste memory bandwidth and DRAM resource. In this paper, we propose Mocha, a non-hierarchical architecture that organizes DRAM and NVM in a flat address space physically, but manages them in a cache/memory hierarchy. Since the commercial NVM device–Intel Optane DC Persistent Memory Modules (DCPMM) actually access the physical media at a granularity of 256 bytes (an Optane block), we manage the DRAM cache at the 256-byte size to adapt to this feature of Optane. This design not only enables fine-grained data migration and management for the DRAM cache, but also avoids write amplification for Intel Optane DCPMM. We also create an Indirect Address Cache (IAC) in Hybrid Memory Controller (HMC) and propose a reverse address mapping table in the DRAM to speed up address translation and cache replacement. Moreover, we exploit a utility-based caching mechanism to filter cold blocks in the NVM, and further improve the efficiency of the DRAM cache. We implement Mocha in an architectural simulator. Experimental results show that Mocha can improve application performance by 8.2% on average (up to 24.6%), reduce 6.9% energy consumption and 25.9% data migration traffic on average, compared with a typical hybrid memory architecture–HSCC.

The cryo-electron microscopy (cryo-EM) is one of the most powerful technologies available today for structural biology. The RELION (Regularized Likelihood Optimization) implements a Bayesian algorithm for cryo-EM structure determination, which is one of the most widely used software in this field. Many researchers have devoted effort to improve the performance of RELION to satisfy the analysis for the ever-increasing volume of datasets. In this paper, we focus on performance analysis of the most time-consuming computation steps in RELION and identify their performance bottlenecks for specific optimizations. We propose several performance optimization strategies to improve the overall performance of RELION, including optimization of expectation step, parallelization of maximization step, accelerating the computation of symmetries, and memory affinity optimization. The experiment results show that our proposed optimizations achieve significant speedups of RELION across representative datasets. In addition, we perform roofline model analysis to understand the effectiveness of our optimizations.

Massive scale of transactions with critical requirements become popular for emerging businesses, especially in E-commerce. One of the most representative applications is the promotional event running on Alibaba’s platform on some special dates, widely expected by global customers. Although we have achieved significant progress in improving the scalability of transactional database systems (OLTP), the presence of contention operations in workloads is still one of the fundamental obstacles to performance improving. The reason is that the overhead of managing conflict transactions with concurrency control mechanisms is proportional to the amount of contentions. As a consequence, generating contented workloads is urgent to evaluate performance of modern OLTP database systems. Though we have kinds of standard benchmarks which provide some ways in simulating contentions, e.g., skew distribution control of transactions, they can not control the generation of contention quantitatively; even worse, the simulation effectiveness of these methods is affected by the scale of data. So in this paper we design a scalable quantitative contention generation method with fine contention granularity control. We conduct a comprehensive set of experiments on popular opensourced DBMSs compared with the latest contention simulation method to demonstrate the effectiveness of our generation work.

Container-based virtualization is becoming increasingly popular in cloud computing due to its efficiency and flexibility. Resource isolation is a fundamental property of containers. Existing works have indicated weak resource isolation could cause significant performance degradation for containerized applications and enhanced resource isolation. However, current studies have almost not discussed the isolation problems of page cache which is a key resource for containers. Containers leverage memory cgroup to control page cache usage. Unfortunately, existing policy introduces two major problems in a container-based environment. First, containers can utilize more memory than limited by their cgroup, effectively breaking memory isolation. Second, the OS kernel has to evict page cache to make space for newly-arrived memory requests, slowing down containerized applications. This paper performs an empirical study of these problems and demonstrates the performance impacts on containerized applications. Then we propose pCache (precise control of page cache) to address the problems by dividing page cache into private and shared and controlling both kinds of page cache separately and precisely. To do so, pCache leverages two new technologies: fair account (f-account) and evict on demand (EoD). F-account splits the shared page cache charging based on per-container share to prevent containers from using memory for free, enhancing memory isolation. And EoD reduces unnecessary page cache evictions to avoid the performance impacts. The evaluation results demonstrate that our system can effectively enhance memory isolation for containers and achieve substantial performance improvement over the original page cache management policy.

Crime risk prediction is helpful for urban safety and citizens’ life quality. However, existing crime studies focused on coarse-grained prediction, and usually failed to capture the dynamics of urban crimes. The key challenge is data sparsity, since that 1) not all crimes have been recorded, and 2) crimes usually occur with low frequency. In this paper, we propose an effective framework to predict fine-grained and dynamic crime risks in each road using heterogeneous urban data. First, to address the issue of unreported crimes, we propose a cross-aggregation soft-impute (CASI) method to deal with possible unreported crimes. Then, we use a novel crime risk measurement to capture the crime dynamics from the perspective of influence propagation, taking into consideration of both time-varying and location-varying risk propagation. Based on the dynamically calculated crime risks, we design contextual features (i.e., POI distributions, taxi mobility, demographic features) from various urban data sources, and propose a zero-inflated negative binomial regression (ZINBR) model to predict future crime risks in roads. The experiments using the real-world data from New York City show that our framework can accurately predict road crime risks, and outperform other baseline methods.

GPUs are widely used in modern high-performance computing systems. To reduce the burden of GPU programmers, operating system and GPU hardware provide great supports for shared virtual memory, which enables GPU and CPU to share the same virtual address space. Unfortunately, the current SIMT execution model of GPU brings great challenges for the virtual-physical address translation on the GPU side, mainly due to the huge number of virtual addresses which are generated simultaneously and the bad locality of these virtual addresses. Thus, the excessive TLB accesses increase the miss ratio of TLB. As an attractive solution, Page Walk Cache (PWC) has received wide attention for its capability of reducing the memory accesses caused by TLB misses.

However, the current PWC mechanism suffers from heavy redundancies, which significantly limits its efficiency. In this paper, we first investigate the facts leading to this issue by evaluating the performance of PWC with typical GPU benchmarks. We find that the repeated L4 and L3 indices of virtual addresses increase the redundancies in PWC, and the low locality of L2 indices causes the low hit ratio in PWC. Based on these observations, we propose a new PWC structure, namely Compressed Page Walk Cache (CPWC), to resolve the redundancy burden in current PWC. Our CPWC can be organized in either direct-mapped mode or set-associated mode. Experimental results show that CPWC increases by 3 times over TPC in the number of page table entries, increases by 38.3% over PWC in L2 index hit ratio and reduces by 26.9% in the memory accesses of page tables. The average memory accesses caused by each TLB miss is reduced to 1.13. Overall, the average IPC can improve by 25.3%.

Blockchain has recently emerged as a research trend, with potential applications in a broad range of industries and context. One particular successful Blockchain technology is smart contract, which is widely used in commercial settings (e.g., high value financial transactions). This, however, has security implications due to the potential to financially benefit froma security incident (e.g., identification and exploitation of a vulnerability in the smart contract or its implementation). Among, Ethereum is the most active and arresting. Hence, in this paper, we systematically review existing research efforts on Ethereum smart contract security, published between 2015 and 2019. Specifically, we focus on how smart contracts can be maliciously exploited and targeted, such as security issues of contract program model, vulnerabilities in the program and safety consideration introduced by program execution environment. We also identify potential research opportunities and future research agenda.

Multiparty private set intersection (PSI) allows several parties, each holding a set of elements, to jointly compute the intersection without leaking any additional information. With the development of cloud computing, PSI has a wide range of applications in privacy protection. However, it is complex to build an efficient and reliable scheme to protect user privacy.

To address this issue, we propose EMPSI, an efficient PSI (with cardinality) protocol in a multiparty setting. EMPSI avoids using heavy cryptographic primitives (mainly rely on symmetric-key encryption) to achieve better performance. In addition, both PSI and PSI with the cardinality of EMPSI are secure against semi-honest adversaries and allow any number of colluding clients (at least one honest client). We also do experiments to compare EMPSI with some state-of-the-art works. The experimental results show that proposed EMPSI(-CA) has better performance and is scalable in the number of clients and the set size.

Cognitive diagnosis is the judgment of the student’s cognitive ability, is a wide-spread concern in educational science. The cognitive diagnosis model (CDM) is an essential method to realize cognitive diagnosis measurement. This paper presents new research on the cognitive diagnosis model and introduces four individual aspects of probability-based CDM and deep learning-based CDM. These four aspects are higher-order latent trait, polytomous responses, polytomous attributes, and multilevel latent traits. The paper also sorts on the contained ideas, model structures and respective characteristics, and provides direction for developing cognitive diagnosis in the future.

Federated learning is a promising learning paradigm that allows collaborative training of models across multiple data owners without sharing their raw datasets. To enhance privacy in federated learning, multi-party computation can be leveraged for secure communication and computation during model training. This survey provides a comprehensive review on how to integrate mainstream multi-party computation techniques into diverse federated learning setups for guaranteed privacy, as well as the corresponding optimization techniques to improve model accuracy and training efficiency. We also pinpoint future directions to deploy federated learning to a wider range of applications.

Music is the language of emotions. In recent years, music emotion recognition has attracted widespread attention in the academic and industrial community since it can be widely used in fields like recommendation systems, automatic music composing, psychotherapy, music visualization, and so on. Especially with the rapid development of artificial intelligence, deep learning-based music emotion recognition is gradually becoming mainstream. This paper gives a detailed survey of music emotion recognition. Starting with some preliminary knowledge of music emotion recognition, this paper first introduces some commonly used evaluation metrics. Then a three-part research framework is put forward. Based on this three-part research framework, the knowledge and algorithms involved in each part are introduced with detailed analysis, including some commonly used datasets, emotion models, feature extraction, and emotion recognition algorithms. After that, the challenging problems and development trends of music emotion recognition technology are proposed, and finally, the whole paper is summarized.

In recent years, the demand for real-time data processing has been increasing, and various stream processing systems have emerged. When the amount of data input to the stream processing system fluctuates, the computing resources required by the stream processing job will also change. The resources used by stream processing jobs need to be adjusted according to load changes, avoiding the waste of computing resources. At present, existing works adjust stream processing jobs based on the assumption that there is a linear relationship between the operator parallelism and operator resource consumption (e.g., throughput), which makes a significant deviation when the operator parallelism increases. This paper proposes a nonlinear model to represent operator performance. We divide the operator performance into three stages, the Non-competition stage, the Non-full competition stage, and the Full competition stage. Using our proposed performance model, given the parallelism of the operator, we can accurately predict the CPU utilization and operator throughput. Evaluated with actual experiments, the prediction error of our model is below 5%. We also propose a quick accurate auto-scaling (QAAS) method that uses the operator performance model to implement the auto-scaling of the operator parallelism of the Flink job. Compared to previous work, QAAS is able to maintain stable job performance under load changes, minimizing the number of job adjustments and reducing data backlogs by 50%.

Instance co-segmentation aims to segment the co-occurrent instances among two images. This task heavily relies on instance-related cues provided by co-peaks, which are generally estimated by exhaustively exploiting all paired candidates in point-to-point patterns. However, such patterns could yield a high number of false-positive co-peaks, resulting in over-segmentation whenever there are mutual occlusions. To tackle with this issue, this paper proposes an instance co-segmentation method via tensor-based salient co-peak search (TSCPS-ICS). The proposed method explores high-order correlations via triple-to-triple matching among feature maps to find reliable co-peaks with the help of co-saliency detection. The proposed method is shown to capture more accurate intra-peaks and inter-peaks among feature maps, reducing the false-positive rate of co-peak search. Upon having accurate co-peaks, one can efficiently infer responses of the targeted instance. Experiments on four benchmark datasets validate the superior performance of the proposed method.

The flourish of deep learning frameworks and hardware platforms has been demanding an efficient compiler that can shield the diversity in both software and hardware in order to provide application portability. Among the existing deep learning compilers, TVM is well known for its efficiency in code generation and optimization across diverse hardware devices. In the meanwhile, the Sunway many-core processor renders itself as a competitive candidate for its attractive computational power in both scientific computing and deep learning workloads. This paper combines the trends in these two directions. Specifically, we propose swTVM that extends the original TVM to support ahead-of-time compilation for architecture requiring cross-compilation such as Sunway. In addition, we leverage the architecture features during the compilation such as core group for massive parallelism, DMA for high bandwidth memory transfer and local device memory for data locality, in order to generate efficient codes for deep learning workloads on Sunway. The experiment results show that the codes generated by swTVM achieve 1.79\times improvement of inference latency on average compared to the state-of-the-art deep learning framework on Sunway, across eight representative benchmarks. This work is the first attempt from the compiler perspective to bridge the gap of deep learning and Sunway processor particularly with productivity and efficiency in mind. We believe this work will encourage more people to embrace the power of deep learning and Sunway many-core processor.

Moving target detection is one of the most basic tasks in computer vision. In conventional wisdom, the problem is solved by iterative optimization under either Matrix Decomposition (MD) or Matrix Factorization (MF) framework. MD utilizes foreground information to facilitate background recovery. MF uses noise-based weights to fine-tune the background. So both noise and foreground information contribute to the recovery of the background. To jointly exploit their advantages, inspired by two framework complementary characteristics, we propose to simultaneously exploit the advantages of these two optimizing approaches in a unified framework called Joint Matrix Decomposition and Factorization (JMDF). To improve background extraction, a fuzzy factorization is designed. The fuzzy membership of the background/foreground association is calculated during the factorization process to distinguish their contributions of both to background estimation. To describe the spatio-temporal continuity of foreground more accurately, we propose to incorporate the first order temporal difference into the group sparsity constraint adaptively. The temporal constraint is adjusted adaptively. Both foreground and the background are jointly estimated through an effective alternate optimization process, and the noise can be modeled with the specific probability distribution. The experimental results of vast real videos illustrate the effectiveness of our method. Compared with the current state-of-the-art technology, our method can usually form the clearer background and extract the more accurate foreground. Anti-noise experiments show the noise robustness of our method.

Uncertain Knowledge Graphs (UKGs) are used to characterize the inherent uncertainty of knowledge and have a richer semantic structure than deterministic knowledge graphs. The research on the embedding of UKG has only recently begun, Uncertain Knowledge Graph Embedding (UKGE) model has a certain effect on solving this problem. However, there are still unresolved issues. On the one hand, when reasoning the confidence of unseen relation facts, the introduced probabilistic soft logic cannot be used to combine multi-path and multi-step global information, leading to information loss. On the other hand, the existing UKG embedding model can only model symmetric relation facts, but the embedding problem of asymmetric relation facts has not be addressed. To address the above issues, a Multiplex Uncertain Knowledge Graph Embedding (MUKGE) model is proposed in this paper. First, to combine multiple information and achieve more accurate results in confidence reasoning, the Uncertain ResourceRank (URR) reasoning algorithm is introduced. Second, the asymmetry in the UKG is defined. To embed asymmetric relation facts of UKG, a multi-relation embedding model is proposed. Finally, experiments are carried out on different datasets via 4 tasks to verify the effectiveness of MUKGE. The results of experiments demonstrate that MUKGE can obtain better overall performance than the baselines, and it helps advance the research on UKG embedding.

Federated learning (FL) has emerged to break data-silo and protect clients’ privacy in the field of artificial intelligence. However, deep leakage from gradient (DLG) attack can fully reconstruct clients’ data from the submitted gradient, which threatens the fundamental privacy of FL. Although cryptology and differential privacy prevent privacy leakage from gradient, they bring negative effect on communication overhead or model performance. Moreover, the original distribution of local gradient has been changed in these schemes, which makes it difficult to defend against adversarial attack. In this paper, we propose a novel federated learning framework with model decomposition, aggregation and assembling (FedDAA), along with a training algorithm, to train federated model, where local gradient is decomposed into multiple blocks and sent to different proxy servers to complete aggregation. To bring better privacy protection performance to FedDAA, an indicator is designed based on image structural similarity to measure privacy leakage under DLG attack and an optimization method is given to protect privacy with the least proxy servers. In addition, we give defense schemes against adversarial attack in FedDAA and design an algorithm to verify the correctness of aggregated results. Experimental results demonstrate that FedDAA can reduce the structural similarity between the reconstructed image and the original image to 0.014 and remain model convergence accuracy as 0.952, thus having the best privacy protection performance and model training effect. More importantly, defense schemes against adversarial attack are compatible with privacy protection in FedDAA and the defense effects are not weaker than those in the traditional FL. Moreover, verification algorithm of aggregation results brings about negligible overhead to FedDAA.

In the past decade, recommender systems have been widely used to provide users with personalized products and services. However, most traditional recommender systems are still facing a challenge in dealing with the huge volume, complexity, and dynamics of information. To tackle this challenge, many studies have been conducted to improve recommender system by integrating deep learning techniques. As an unsupervised deep learning method, autoencoder has been widely used for its excellent performance in data dimensionality reduction, feature extraction, and data reconstruction. Meanwhile, recent researches have shown the high efficiency of autoencoder in information retrieval and recommendation tasks. Applying autoencoder on recommender systems would improve the quality of recommendations due to its better understanding of users’ demands and characteristics of items. This paper reviews the recent researches on autoencoder-based recommender systems. The differences between autoencoder-based recommender systems and traditional recommender systems are presented in this paper. At last, some potential research directions of autoencoder-based recommender systems are discussed.

Most current crowdsourced logistics aim to minimize systems cost and maximize delivery capacity, but the efforts of crowdsourcers such as drivers are almost ignored. In the delivery process, drivers usually need to take long-distance detours in hitchhiking rides based package deliveries. In this paper, we propose an approach that integrates offline trajectory data mining and online route-and-schedule optimization in the hitchhiking ride scenario to find optimal delivery routes for packages and drivers. Specifically, we propose a two-phase framework for the delivery route planning and scheduling. In the first phase, the historical trajectory data are mined offline to build the package transport network. In the second phase, we model the delivery route planning and package-taxi matching as an integer linear programming problem and solve it with the Gurobi optimizer. After that, taxis are scheduled to deliver packages with optimal delivery paths via a newly designed scheduling strategy. We evaluate our approach with the real-world datasets; the results show that our proposed approach can complete citywide package deliveries with a high success rate and low extra efforts of taxi drivers.

A threshold signature is a special digital signature in which the N-signer share the private key x and can construct a valid signature for any subset of the included t-signer, but less than t-signer cannot obtain any information. Considering the breakthrough achievements of threshold ECDSA signature and threshold Schnorr signature, the existing threshold SM2 signature is still limited to two parties or based on the honest majority setting, there is no more effective solution for the multiparty case. To make the SM2 signature have more flexible application scenarios, promote the application of the SM2 signature scheme in the blockchain system and secure cryptocurrency wallets. This paper designs a non-interactive threshold SM2 signature scheme based on partially homomorphic encryption and zero-knowledge proof. Only the last round requires the message input, so make our scheme non-interactive, and the pre-signing process takes 2 rounds of communication to complete after the key generation. We allow arbitrary threshold t\le n and design a key update strategy. It can achieve security with identifiable abort under the malicious majority, which means that if the signature process fails, we can find the failed party. Performance analysis shows that the computation and communication costs of the pre-signing process grows linearly with the parties, and it is only 1/3 of the Canetti’s threshold ECDSA (CCS'20).