The hidden Markov models are statistical models used in many real-world applications and communities. The use of hidden Markov models has become predominant in the last decades, as evidenced by a large number of published papers. In this survey, 146 papers (101 from Journals and 45 from Conferences/Workshops) from 93 Journals and 44 Conferences/Workshops are considered. The authors evaluate the literature based on hidden Markov model variants that have been applied to various application fields. The paper represents a short but comprehensive description of research on hidden Markov model and its variants for various applications. The paper shows the significant trends in the research on hidden Markov model variants and their applications.

A Systematic Review of Hidden Markov Models and Their Applications

Recently, there has been significant progress in teaching language models to perform step-by-step reasoning to solve complex numerical reasoning tasks. Chain-of-thoughts prompting (CoT) is by far the state-of-art method for these tasks. CoT uses language models to perform both reasoning and computation in the multi-step `thought' process. To disentangle computation from reasoning, we propose `Program of Thoughts' (PoT), which uses language models (mainly Codex) to express the reasoning process as a program. The computation is relegated to an external computer, which executes the generated programs to derive the answer. We evaluate PoT on five math word problem datasets (GSM, AQuA, SVAMP, TabMWP, MultiArith) and three financial-QA datasets (FinQA, ConvFinQA, TATQA) for both few-shot and zero-shot setups. Under both few-shot and zero-shot settings, PoT can show an average performance gain over CoT by around 12\% across all the evaluated datasets. By combining PoT with self-consistency decoding, we can achieve SoTA performance on all math problem datasets and near-SoTA performance on financial datasets. All of our data and code are released in Github\footnote{\url{https://github.com/wenhuchen/Program-of-Thoughts}}.

Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks

Nowadays, numerous corporations (such as Google, Baidu, etc.) require an efficient and effective search algorithm to crawl out the images with queried objects from databases. Moreover, privacy protection is a significant issue such that confidential images must be encrypted in corporations. Nevertheless, decrypting and then classifying millions of encrypted images becomes a heavy burden to computation. In this paper, we proposed an encrypted image classification framework based on multi-layer extreme learning machine that is able to directly classify encrypted images without decryption. Experiments were conducted on popular handwritten digits and letters databases. Results demonstrate that the proposed framework is secure, efficient and accurate for classifying encrypted images.

Encrypted image classification based on multilayer extreme learning machine

Knowledge about a network protocol to understand the communication between entities is necessary for vulnerability research, penetration testing, malware analysis, network reconnaissance, and network modeling Traffic analysis is one approach to infer a protocol, and this approach has specific challenges, tasks, methods, and solutions In this survey, we collect tools presented by prior research in the field of protocol reverse engineering by static traffic trace analysis We dissect each tool to discern the individual mechanisms and the algorithms on which they are based, then categorize and contrast the mechanisms and algorithms used in static traffic trace analysis to discuss how successfully they were applied in each case To structure our discussion about the tools, we compared classification schemes for protocol reverse engineering We present and discuss an explicit process model for static traffic trace analysis to reveal the common structure of the decomposed tools and frameworks from previous research Via discussions of the algorithms applied within each tool, we show relations between tools, methods, and the process for each process task We validate our model by applying it to each of the tools, then provide an outline of the utility of protocol reverse engineering Beginning with the process description, we deduce which solutions and algorithms have already been investigated and where challenges remain to determine how new solutions may be researched in the future Across the entire field of protocol reverse engineering, few implementations of tools and frameworks are publicly available, which remains a prevalent problem

/pdf/survey-of-protocol-reverse-engineering-algorithms-27adlijiuu.pdf

Survey of Protocol Reverse Engineering Algorithms: Decomposition of Tools for Static Traffic Analysis

Remote sensing scene classification has gained increasing interest in remote sensing image understanding and feature representation is the crucial factor for classification methods. Convolutional N...

Remote sensing scene classification based on high-order graph convolutional network

Deep convolutional neural networks show great advantages in computer vision tasks, such as image classification and object detection. However, the networks have complex network structure which include a large number of layers such as convolutional layers and pooling layers. They greatly consume valuable computing and memory resources, and also hugely waste training time. Therefore, we propose a novel shallow convolutional neural network (SCNNB) to overcome the above limitations for image classification, which uses batch normalization techniques to accelerate training convergence and improve the accuracy. The SCNNB network has only 4 layers with small size of convolution kernels, which requires low time complexity and space complexity. In the experiments, we compare the SCNNB model with two variant models and the classical SCNN model on the two benchmark image datasets. Experimental results show that compared to SCNN model, the SCNNB model can quickly learn the features of the data and achieve the highest classification accuracy of 93.69% with 3.8 M time complexity on fashion-MNIST.

/pdf/shallow-convolutional-neural-network-for-image-8twgtut756.pdf

Shallow convolutional neural network for image classification

With the rapid development of Internet, especially the mobile Internet, the new applications or network attacks emerge in a high rate in recent years. More and more traffic becomes unknown due to the lack of protocol specifications about the newly emerging applications. Automatic protocol reverse engineering is a promising solution for understanding this unknown traffic and recovering its protocol specification. One challenge of protocol reverse engineering is to determine the length of protocol keywords and message fields. Existing algorithms are designed to select the longest substrings as protocol keywords, which is an empirical way to decide the length of protocol keywords. In this paper, we propose a novel approach to determine the optimal length of protocol keywords and recover message formats of Internet protocols by maximizing the likelihood probability of message segmentation and keyword selection. A hidden semi-Markov model is presented to model the protocol message format. An affinity propagation mechanism based clustering technique is introduced to determine the message type. The proposed method is applied to identify network traffic and compare the results with existing algorithm.

/pdf/analyzing-network-protocols-of-application-layer-using-4q9wg9gpuj.pdf

Analyzing Network Protocols of Application Layer Using Hidden Semi-Markov Model

Wireless Sensor Networks (WSNs) have a wide range of applications scenarios in computer vision, from pedestrian detection to robotic visual navigation. In response to the growing visual data services in WSNs, we propose a proactive caching strategy based on Stacked Sparse Autoencoder (SSAE) to predict content popularity (PCDS2AW). Firstly, based on Software Defined Network (SDN) and Network Function Virtualization (NFV) technologies, a distributed deep learning network SSAE is constructed in the sink nodes and control nodes of the WSN network. Then, the SSAE network structure parameters and network model parameters are optimized through training. The proactive cache strategy implementation procedure is divided into four steps. The SDN controller is responsible for dynamically collecting user request data package information in the WSNs network. The SSAEs predicts the packet popularity based on the SDN controller obtaining user request data. The SDN controller generates a corresponding proactive cache strategy according to the popularity prediction result. Implement the proactive caching strategy at the WSNs cache node. In the simulation, we compare the influence of spatiotemporal data on the SSAE network structure. Compared with the classic caching strategy Hash + LRU, Betw + LRU, and classic prediction algorithms SVM and BPNN, the proposed PCDS2AW proactive caching strategy can significantly improve WSN performance.

/pdf/deep-learning-based-proactive-caching-for-effective-wsn-4v3ogmmibf.pdf

Deep Learning Based Proactive Caching for Effective WSN-Enabled Vision Applications

With higher-order neighborhood information of graph network, the accuracy of graph representation learning classification can be significantly improved. However, the current higher order graph convolutional network has a large number of parameters and high computational complexity. Therefore, we propose a Hybrid Lower order and Higher order Graph convolutional networks (HLHG) learning model, which uses weight sharing mechanism to reduce the number of network parameters. To reduce computational complexity, we propose a novel fusion pooling layer to combine the neighborhood information of high order and low order. Theoretically, we compare the model complexity of the proposed model with the other state-of-the-art model. Experimentally, we verify the proposed model on the large-scale text network datasets by supervised learning, and on the citation network datasets by semi-supervised learning. The experimental results show that the proposed model achieves highest classification accuracy with a small set of trainable weight parameters.

Hybrid Low-order and Higher-order Graph Convolutional Networks

With the higher-order neighborhood information of a graph network, the accuracy of graph representation learning classification can be significantly improved However, the current higher-order graph convolutional networks have a large number of parameters and high computational complexity Therefore, we propose a hybrid lower-order and higher-order graph convolutional network (HLHG) learning model, which uses a weight sharing mechanism to reduce the number of network parameters To reduce the computational complexity, we propose a novel information fusion pooling layer to combine the high-order and low-order neighborhood matrix information We theoretically compare the computational complexity and the number of parameters of the proposed model with those of the other state-of-the-art models Experimentally, we verify the proposed model on large-scale text network datasets using supervised learning and on citation network datasets using semisupervised learning The experimental results show that the proposed model achieves higher classification accuracy with a small set of trainable weight parameters

/pdf/hybrid-low-order-and-higher-order-graph-convolutional-7ga9qdg7vq.pdf

Fangyuan Lei

Papers

Shallow convolutional neural network for image classification

Analyzing Network Protocols of Application Layer Using Hidden Semi-Markov Model

Deep Learning Based Proactive Caching for Effective WSN-Enabled Vision Applications

Hybrid Low-order and Higher-order Graph Convolutional Networks

Hybrid Low-Order and Higher-Order Graph Convolutional Networks.