| Neural network and deep learning system, learns weights of output neurons such that network can generate specific temporal patterns, where connectivity and weights of hidden neurons have memory and are fixed and randomly assigned | |
| 2023-10-16 | |
| 专利权人 | BHARATH HIGHER EDUCATION & RES INST (BHAR-Non-standard) |
| 申请日期 | 2023-10-16 |
| 专利号 | IN202341069682-A |
| 成果简介 | NOVELTY - The system learns weights of output neurons such that a network can generate specific temporal patterns, where connectivity and weights of hidden neurons have memory and are fixed and randomly assigned. DRNs take the deep learning world by storm when deep residual learning is released for image recognition. The networks led to first-place winning entries in all five main tracks of the Image Net, which covered image classification, object detection, and semantic segmentation. USE - Neural network and deep learning system. ADVANTAGE - The system improves performance of deep convolutional neural networks (DCNNs) in computer vision by two critical components such as continued growth of computational power such that large labeled data sets take advantage of power of a deep multi-layer architecture. |
| IPC 分类号 | G06N-020/00 ; G06N-003/00 ; G06N-003/04 ; G06N-003/08 ; G06N-007/00 |
| 国家 | 印度 |
| 专业领域 | 信息技术 |
| 语种 | 英语 |
| 成果类型 | 专利 |
| 文献类型 | 科技成果 |
| 条目标识符 | http://119.78.100.226:8889/handle/3KE4DYBR/19766 |
| 专题 | 中国科学院新疆生态与地理研究所 |
| 作者单位 | BHARATH HIGHER EDUCATION & RES INST (BHAR-Non-standard) |
| 推荐引用方式 GB/T 7714 | MUTHUKUMARAVEL A,CHIDAMBRAM K,NAVEENCHANDRAN P,et al. Neural network and deep learning system, learns weights of output neurons such that network can generate specific temporal patterns, where connectivity and weights of hidden neurons have memory and are fixed and randomly assigned. IN202341069682-A[P]. 2023. |
| 条目包含的文件 | 条目无相关文件。 | |||||
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论