{"title":"DAVO:A Monocular Visual Odometry Method Based on Dual Attention","authors":"Jiahao Li, Bin Zheng","doi":"10.1145/3579654.3579704","DOIUrl":"https://doi.org/10.1145/3579654.3579704","url":null,"abstract":"In recent years, Visual odometry(VO) has been widely used in fields such as autonomous driving and virtual reality. Traditional methods for solving visual odometry rely on complex processes such as feature extraction, feature matching and camera calibration, and have low robustness and serious accuracy deficiency problems in challenging environments. In this paper, we propose a dual attention monocular visual odometry model that integrates Deep Learning(DL) with Reinforcement Learning(RL), named DAVO (Dual Attention Visual Odometry). The model combines a recurrent attention network model with a self-attentive mechanism to solve the relative poses of six degrees of freedom(6-DoF) by learning the image region locations that are favorable for the model pose estimation through a reinforcement learning algorithm. Finally, the model is evaluated and compared on the publicly available dataset KITTI. Compared with other mainstream models, DAVO only inputs 14.04% of the data in the image preprocessing stage, runs faster and outperforms most of the mainstream models.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121366043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Design of High-performance SoC Simulation Model Based on Verilator","authors":"Yuan Chi, Xian Lin, Xin Zheng","doi":"10.1145/3579654.3579751","DOIUrl":"https://doi.org/10.1145/3579654.3579751","url":null,"abstract":"With the development of Moore's Law, the design complexity of SoC is increasing, and the traditional simulation verification method has been unable to meet the rapid iteration of products. A high-performance SoC simulation model design scheme based on Verilator is proposed to improve the speed of SoC simulation verification and shorten the chip development and design cycle. Based on the ESL design method, Openc910 is selected as the starting point to design a high-performance SoC simulation model based on Verilator. The SM3 and SM4 crypto modules are designed in the SoC, and Iverilog simulation tool is used for comparison. Compared with using the traditional Iverilog simulation, the experimental results using Verilator show the effectiveness of the design.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114606479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Cross-Individual Obstructive Obstructive Apnea Detection in Snoring Signals Using Hybrid Deep Neural Networks","authors":"Xu Lin, Yun Lu, Heng Li, Yukun Qian, Lianyu Zhou, Mingjiang Wang","doi":"10.1145/3579654.3579670","DOIUrl":"https://doi.org/10.1145/3579654.3579670","url":null,"abstract":"Sleep apnea syndrome (SAS) is a common sleep problem, among which obstructive sleep apnea (OSA) is the most common. It is estimated that 936 million adults aged 30-69 years suffer from mild to severe obstructive sleep apnea that can result in poor sleep quality and even endanger their lives. In our study, 2051 OSA snoring fragments and 2271 normal snoring fragments were collected, and then the two were classified by the hybrid neural network. The most important innovation of this paper is the cross-individual snoring classification, which is different from the previous work, making the model more generalized. The experimental dataset was from 24 patients, the snores of 20 patients were used for the training model, and the snores of 4 people were used for the test. Finally, the accuracy of classification on the test set was 73.75%, and a portable snore classification platform is realized by using an embedded platform and edge computing.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122402224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Part-GCNet: Partitioning Graph Convolutional Network for Multi-Label Recognition","authors":"Yuan Zhang, Tao Han, Bing Wei, K. Hao","doi":"10.1145/3579654.3579659","DOIUrl":"https://doi.org/10.1145/3579654.3579659","url":null,"abstract":"During the rapid development of deep learning, the multi-label recognition task has achieved pretty performance. Recently, the emergence of graph convolution network (GCN) has further improved the accuracy of multi-label recognition. However, in the learning process, how to better represent the feature information of labels and innovatively design structures to obtain good recognition performance is still unclear. To solve these problems, we propose a partitioning graph convolutional network framework for multi-label recognition. First, we segregate the computational graph into multiple sub-graphs. Then, we perform batch normalization operation on each output layer, which can further improve the recognition performance of the network. Finally, extensive experiments are carried out on a multi-label PPT dataset, showing that our proposed solution can greatly improve the feature information utilization of labels and improve the recognition performance.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130874644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Offline Imitation Learning Using Reward-free Exploratory Data","authors":"Hao Wang, Dawei Feng, Bo Ding, W. Li","doi":"10.1145/3579654.3579753","DOIUrl":"https://doi.org/10.1145/3579654.3579753","url":null,"abstract":"Offline imitative learning(OIL) is often used to solve complex continuous decision-making tasks. For these tasks such as robot control, automatic driving and etc., it is either difficult to design an effective reward for learning or very expensive and time-consuming for agents to collect data interactively with the environment. However, the data used in previous OIL methods are all gathered by reinforcement learning algorithms guided by task-specific rewards, which is not a true reward-free premise and still suffers from the problem of designing an effective reward function in real tasks. To this end, we propose the reward-free exploratory data driven offline imitation learning (ExDOIL) framework. ExDOIL first trains an unsupervised reinforcement learning agent by interacting with the environment, and collects enough unsupervised exploration data during training; Then, a task independent yet simple and efficient reward function is used to relabel the collected data; Finally, an agent is trained to imitate the expert to complete the task through a conventional RL algorithm such as TD3. Extensive experiments on continuous control tasks demonstrate that the proposed framework can achieve better imitation performance(28% higher episode returns on average) comparing with previous SOTA method(ORIL) without any task-specific rewards.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121502106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Design of Soil Water-saving Irrigation Control System Based on Fuzzy Control","authors":"Xiaojie Tang, Yuanlin Li, Chengfen Jia","doi":"10.1145/3579654.3579743","DOIUrl":"https://doi.org/10.1145/3579654.3579743","url":null,"abstract":"The traditional crop irrigation water utilization rate is low, it can not be very good water saving.In order to save water resources, a soil water-saving irrigation control system based on fuzzy control algorithm was proposed with the large cherry planting in Dalian, China as an example. The overall system structure is designed, the hardware circuit is built and the corresponding components are selected. At the same time, the fuzzy logic is dissolved into the algorithm, and the fuzzy controller of soil optimum humidity, the fuzzy controller of irrigation time point suitability and the fuzzy controller of irrigation time are designed. The fuzzy controller is used to calculate the irrigation time of soil. The simulation findings demonstrate that this design can automatically adjust the irrigation duration based on the crop environment, the current soil humidity differential, and the crop growth period to increase crop yield and conserve water.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121782479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Cedrique Rovile Njieutcheu Tassi, A. Börner, Rudolph Triebel
{"title":"Regularization Strength Impact on Neural Network Ensembles","authors":"Cedrique Rovile Njieutcheu Tassi, A. Börner, Rudolph Triebel","doi":"10.1145/3579654.3579661","DOIUrl":"https://doi.org/10.1145/3579654.3579661","url":null,"abstract":"In the last decade, several approaches have been proposed for regularizing deeper and wider neural networks (NNs), which is of importance in areas like image classification. It is now common practice to incorporate several regularization approaches in the training procedure of NNs. However, the impact of regularization strength on the properties of an ensemble of NNs remains unclear. For this reason, the study empirically compared the impact of NNs built based on two different regularization strengths (weak regularization (WR) and strong regularization (SR)) on the properties of an ensemble, such as the magnitude of logits, classification accuracy, calibration error, and ability to separate true predictions (TPs) and false predictions (FPs). The comparison was based on results from different experiments conducted on three different models, datasets, and architectures. Experimental results show that the increase in regularization strength 1) reduces the magnitude of logits; 2) can increase or decrease the classification accuracy depending on the dataset and/or architecture; 3) increases the calibration error; and 4) can improve or harm the separability between TPs and FPs depending on the dataset, architecture, model type and/or FP type.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129219168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Adversarial Training on Weights for Graph Neural Networks","authors":"Hao Xue, Xin Wang, Ying Wang","doi":"10.1145/3579654.3579738","DOIUrl":"https://doi.org/10.1145/3579654.3579738","url":null,"abstract":"Despite the fact that Graph Neural Networks (GNNs) have been extensively used for graph embedding representation, it is challenging to train well-performing GNNs on graphs with good generalization due to the limitation of overfitting. Previous research in Computer Vision (CV) has shown that the lack of generalization usually corresponds to the convergence of model parameters to sharp local minima. However, there is still a lack of related research in the field of graph analysis. In this paper, we investigate the loss landscape of models from the weight change perspective and show that the vanilla training method tends to cause GNNs to fall into sharp local minima with poor generalization. To tackle this problem, we propose a method named Adversarial Training on Weights (ATW) to flatten the weight loss landscape using adversarial training, thus improving the generalization of GNNs. Extensive experiments with multiple backbones on various datasets demonstrate the effectiveness of our method.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127763522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Time interval-aware graph with self-attention for sequential recommendation","authors":"Zhuo Chen, Weiwei Wang","doi":"10.1145/3579654.3579729","DOIUrl":"https://doi.org/10.1145/3579654.3579729","url":null,"abstract":"Sequential recommendation, as a branch under the recommendation system, obtains the user’s interest changes from the user’s interaction history to predict the next item. The neural network structure, Transformer and Graph Neural Networks (GNN) have been widely used in recommendation systems due to their ability to represent sequences and capture high-order information. However, previous models only rank actions in the time order of occurrence, ignoring the effect of the time interval between adjacent actions, which usually reflects the user’s preferences. To fully use time information, we design the Time Interval-aware Graph with Self-attention for sequential recommendation (TIGSA). Specifically, we first construct a time interval-aware graph, which integrates the information of different time intervals in all user action sequences. The time interval of two items determines the weight of each edge in the graph. Then the item model combined with the time interval information is obtained through the Graph Convolutional Networks (GCN). Finally, the self-attention block is used to adaptively compute the attention weights of the items in the sequence. Experiments show that our method outperforms other recommendation models on three public datasets and different evaluation metrics.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121196466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"VSLayout: Visual-Semantic Representation Learning For Document Layout Analysis","authors":"Shan Wang, Jing Jiang, Yanjun Jiang, Xuesong Zhang","doi":"10.1145/3579654.3579767","DOIUrl":"https://doi.org/10.1145/3579654.3579767","url":null,"abstract":"Document layout analysis (DLA), aiming to extract and classify the structural regions, is a rather challenging and critical step for many downstream document understanding tasks. Although the fusion of text (semantics) and image (vision) features has shown significant advantages for DLA, existing methods either require simultaneous text-image pair inputs, which is not applicable when only document images are available, or have to resort to an optical character recognition (OCR) preprocessing. This paper learns the visual-sematic representation for DLA only from the imaging modality of documents, which greatly extends the applicability of DLA to practical applications. Our method consists of three phases. Firstly, we train a text feature extractor (TFE) for document images via cross-modal supervision that enforces the coherence between the outputs of TFE and the text embedding map generated by Sent2Vec. Then the pretrained TFE gets further adapted using only the document images and extracts shallow semantic features that will be further fed into the third stage. Finally, a two-stream network is employed to extract the deep semantic and visual features, and their fusion is used as the input to a detector module, e.g., the RPN (Region Proposal Network), to generate the final results. On benchmark datasets, we demonstrate that the proposed TFE model outperforms main-stream semantic embedding counterparts and that our approach achieves superior DLA performance to baseline methods.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"336 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115880248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}