Pub Date : 2023-11-10DOI: 10.1142/s2972335324500017
Elahe Vahdani, Longlong Jing, Matt Huenerfauth, Yingli Tian
{"title":"Multi-Modal Multi-Channel American Sign Language Recognition","authors":"Elahe Vahdani, Longlong Jing, Matt Huenerfauth, Yingli Tian","doi":"10.1142/s2972335324500017","DOIUrl":"https://doi.org/10.1142/s2972335324500017","url":null,"abstract":"","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":" July","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135186544","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}
Pub Date : 2023-10-27DOI: 10.1142/s2972335323500011
Lin Zhang, Ling Feng, Kan Chen, Choy Heng Lai
It has been recently demonstrated that optimal neural networks operate near the asymptotic edge of chaos for state-of-the-art feed-forward neural networks, where its generalization power is maximal due to the highest number of asymptotic metastable states. However, how to leverage this principle to improve the model training process remains open. Here, by mapping the model evolution during training to the phase diagram in the classic analytic result of Sherrington–Kirkpatrick model in spin glasses, we illustrate on a simple neural network model that one can provide principled training of the network without manually tuning the training hyper-parameters. In particular, we provide a semi-analytical method to set the optimal weight decay strength, such that the model will converge toward the edge of chaos during training. Consequently, such hyper-parameter setting leads the model to achieve the highest test accuracy. Another benefit for restricting the model at the edge of chaos is its robustness against the common practical problem of label noise, as we find that it automatically avoids fitting the shuffled labels in the training samples while maintaining good fitting to the correct labels, providing simple means of achieving good performance on noisy labels without any additional treatment.
{"title":"Asymptotic Edge of Chaos as Guiding Principle for Neural Network Training","authors":"Lin Zhang, Ling Feng, Kan Chen, Choy Heng Lai","doi":"10.1142/s2972335323500011","DOIUrl":"https://doi.org/10.1142/s2972335323500011","url":null,"abstract":"It has been recently demonstrated that optimal neural networks operate near the asymptotic edge of chaos for state-of-the-art feed-forward neural networks, where its generalization power is maximal due to the highest number of asymptotic metastable states. However, how to leverage this principle to improve the model training process remains open. Here, by mapping the model evolution during training to the phase diagram in the classic analytic result of Sherrington–Kirkpatrick model in spin glasses, we illustrate on a simple neural network model that one can provide principled training of the network without manually tuning the training hyper-parameters. In particular, we provide a semi-analytical method to set the optimal weight decay strength, such that the model will converge toward the edge of chaos during training. Consequently, such hyper-parameter setting leads the model to achieve the highest test accuracy. Another benefit for restricting the model at the edge of chaos is its robustness against the common practical problem of label noise, as we find that it automatically avoids fitting the shuffled labels in the training samples while maintaining good fitting to the correct labels, providing simple means of achieving good performance on noisy labels without any additional treatment.","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":"268 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136233157","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}
Pub Date : 2023-09-08DOI: 10.1142/s2972335323020015
Bud Mishra
{"title":"AI, Thinking Machines and A Vast Active Living Intelligent System","authors":"Bud Mishra","doi":"10.1142/s2972335323020015","DOIUrl":"https://doi.org/10.1142/s2972335323020015","url":null,"abstract":"","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87302307","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}
Pub Date : 2023-01-01DOI: 10.12677/airr.2023.123028
恩兴 赵
The recognition of traffic lights is crucial for driver assistance systems, which can help reduce acci-赵恩兴，王超
{"title":"Traffic Signals Recognition Based on YOLOv8","authors":"恩兴 赵","doi":"10.12677/airr.2023.123028","DOIUrl":"https://doi.org/10.12677/airr.2023.123028","url":null,"abstract":"The recognition of traffic lights is crucial for driver assistance systems, which can help reduce acci-赵恩兴，王超","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":"46 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73516724","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}
Pub Date : 2023-01-01DOI: 10.12677/airr.2023.122012
灶荣 黄
{"title":"Recognition and Classification Technology of Abnormal Gait in Stroke Patients","authors":"灶荣 黄","doi":"10.12677/airr.2023.122012","DOIUrl":"https://doi.org/10.12677/airr.2023.122012","url":null,"abstract":"","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78637138","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}
Pub Date : 2023-01-01DOI: 10.12677/airr.2023.123023
冰 陈
This paper presents a crowd social distancing detection method based on the YOLO model, aiming to effectively control crowd movement and prevent the transmission of infectious diseases in densely populated public areas. The proposed method leverages the YOLO model to perform pedestrian detection on videos captured in public areas, thereby extracting precise positional key
{"title":"A Social Distancing Detection Method for Public Area Crowds Based on YOLO","authors":"冰 陈","doi":"10.12677/airr.2023.123023","DOIUrl":"https://doi.org/10.12677/airr.2023.123023","url":null,"abstract":"This paper presents a crowd social distancing detection method based on the YOLO model, aiming to effectively control crowd movement and prevent the transmission of infectious diseases in densely populated public areas. The proposed method leverages the YOLO model to perform pedestrian detection on videos captured in public areas, thereby extracting precise positional key","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":"2003 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87056458","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}
Pub Date : 2023-01-01DOI: 10.12677/airr.2023.124031
苗苗 苟
{"title":"Knowledge Base Question Answering Method Based on Semantic Fusion","authors":"苗苗 苟","doi":"10.12677/airr.2023.124031","DOIUrl":"https://doi.org/10.12677/airr.2023.124031","url":null,"abstract":"","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135668110","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}
Pub Date : 2023-01-01DOI: 10.12677/airr.2023.122014
博文 范
{"title":"Domain Specific Batch Normalization Based on Adversarial Domain Adaptation Image Classification","authors":"博文 范","doi":"10.12677/airr.2023.122014","DOIUrl":"https://doi.org/10.12677/airr.2023.122014","url":null,"abstract":"","PeriodicalId":68167,"journal":{"name":"人工智能与机器人研究","volume":"26 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74094882","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：