相关实验视频
Updated: Sep 11, 2025
13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
9.3K
基于可差分的神经架构搜索改进模型的图像分类优化技术
Yuxuan Ji1, Wenshu Li1, Nan Yu1
1College of Information Engineering, Shanghai Maritime University, Shanghai, China.
PloS one
|August 13, 2025
概括
这项研究通过改进神经架构搜索 (NAS) 以视觉注意力和残余结构来增强图像分类. 新型号实现了更高的准确性和效率,克服了捕获本地和远程信息的局限性.
科学领域:
- 计算机视觉 计算机视觉
- 人工智能的人工智能
- 机器学习 机器学习
背景情况:
- 微分神经架构搜索 (NAS) 在捕获本地和远距离信息方面存在局限性,影响图像分类的准确性.
- 现有的NAS方法因局部信息获取弱而困难,阻碍了复杂的视觉任务的性能.
研究的目的:
- 引入视觉注意力机制和改进的模型架构,以提高NAS中的信息获取和分类准确性.
- 解决传统可差分NAS在捕获本地和远程视觉细节方面的局限性.
主要方法:
- 提出了一种改进的模型,包括视觉注意力机制.
- 取代了原来的卷积运算符,并在宏观架构中添加了剩余结构.
- 在CIFAR-10,CIFAR-100和ImageNet数据集上训练和测试模型.
主要成果:
- 在600轮培训后,在CIFAR-10上获得了97.2%的准确性.
- 在CIFAR-100的运行时内存使用率降低到44.52% (减少44.56%).
- 在ImageNet上,获得了94.01%的分类准确度,搜索参数为4.8MB,搜索时间为0.5d,FLOP为3.7G.
结论:
- 改进的模型有效地克服了在本地和远程信息采集方面传统的NAS缺陷.
- 这项研究为提高图像分类的准确性和效率提供了重要的技术支持.
- 与其他主流算法相比,拟议的方法表现出优越的性能.
相关概念视频
02:07Improving Translational Accuracy
11.9K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
11.9K
01:31Classification of Systems-II
241
Continuous-time systems have continuous input and output signals, with time measured continuously. These systems are generally defined by differential or algebraic equations. For instance, in an RC circuit, the relationship between input and output voltage is expressed through a differential equation derived from Ohm's law and the capacitor relation,
241
01:26Classification of Systems-I
301
Linearity is a system property characterized by a direct input-output relationship, combining homogeneity and additivity.
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
301
01:18Reducing Line Loss
193
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
193