混合细粒度分布估计为少数人学习:统计数据从类别和实例转移
概括
混合细粒度分布估计 (HGDE) 通过结合类别和实例统计数据来改进少量学习. 这种新的方法增强了样本的代表性,并提高了一些射击任务的准确性.
科学领域:
- 机器学习 机器学习
- 计算机视觉 计算机视觉
背景情况:
- 短暂的学习 (FSL) 面临着由于数据稀缺的挑战.
- 在FSL中对类别级分布的估计可能会导致低于最佳的性能,因为基础类别和新型类别之间的差异.
研究的目的:
- 引入混合粒度分布估计 (HGDE) 以更有效地估计FSL的分布.
- 通过整合粗粒度和细粒度统计数据来增强新类别的特征.
主要方法:
- HGDE将类别级统计数据与来自最近基样的实例级统计数据集成在一起.
- 统计数据采用线性插值进行融合,以创建新类别的稳健分布.
- 采用了精细的估计技术,包括用于共变量的平均值和主要组件保留的加权总和.
主要成果:
- 在四个FSL基准 (Mini-ImageNet, Tiered-ImageNet, CUB, CIFAR-FS) 中,HGDE展示了有效的分布估计能力.
- 观察到显著的准确性增长,在CUB.上的一次性任务中有超过1.8%的改进.
- 该方法有效地平衡了平均精度和差异多样性.
结论:
- 通过改进分布估计,HGDE提供了一种多功能和有效的解决方案,用于少量学习.
- 混合方法捕捉了细微的特征,仅仅通过类别级估计而忽视的特征.
- HGDE增强了样本的多样性和代表性,从而提高了FSL的性能.
更多相关视频
08:05Measuring Statistical Learning Across Modalities and Domains in School-Aged Children Via an Online Platform and Neuroimaging Techniques
Published on: June 30, 2020
8.1K
08:45Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
Published on: October 24, 2012
15.3K
相关概念视频
Distributions to Estimate Population Parameter
5.2K
The accurate values of population parameters such as population proportion, population mean, and population standard deviation (or variance) are usually unknown. These are fixed values that can only be estimated from the data collected from the samples. The estimates of each of these parameters are sample proportion, the sample mean, and sample standard deviation (or variance). To obtain the values of these sample statistics, data are required that have particular distribution and central...
5.2K
Statistical Significance
22.2K
Once data is collected from both the experimental and the control groups, a statistical analysis is conducted to find out if there are meaningful differences between the two groups. A statistical analysis determines how likely any difference found is due to chance (and thus not meaningful). In psychology, group differences are considered meaningful, or significant, if the odds that these differences occurred by chance alone are 5 percent or less. Stated another way, if we repeated this...
22.2K
Hybrid Zones
22.0K
Hybrid zones are narrow regions where two closely related species interact, mate, and produce hybrids. Relative to either parent species, hybrids may possess distinct phenotypic or genetic differences that impact their survival and reproductive success. The genetic variances introduced by hybridization influence species diversity and speciation processes within the hybrid zone.
22.0K
Categories of Equilibrium
5.7K
Equilibrium is a crucial concept in physics, enabling us to understand how forces interact with bodies to produce no or constant motion. In two-dimensional equilibrium, force systems can be classified into different categories based on their characteristics.
One of the categories of equilibrium is collinear equilibrium, which involves forces acting along a straight line. This type of equilibrium requires only one force equation in the direction of the forces, as the other equations are...
One of the categories of equilibrium is collinear equilibrium, which involves forces acting along a straight line. This type of equilibrium requires only one force equation in the direction of the forces, as the other equations are...
5.7K
Drug Classes and Categories
3.1K
Drugs can be classified according to their chemical composition or their intended therapeutic application. For instance, anti-infective agents that possess the ability to eliminate pathogens or suppress their growth and reproduction can be grouped based on the organisms they target or their chemical structure. Furthermore, drugs can be divided into prescription, nonprescription, or controlled substances. Prescription medications, such as antibiotics, require oversight from a licensed healthcare...
3.1K
Probability in Statistics
23.5K
Probability is the likelihood of an event occurring. The term event is defined as a collection of results of a procedure. An event is a simple event when an outcome cannot be divided into simpler parts.
An example of a simple event is a coin toss. The result of a coin toss is either a head or a tail. Here, head and tail are two simple events. These two simple events make up the sample space. Further, the probability of an event occurring falls within the range of 0 to 1. The probability of an...
An example of a simple event is a coin toss. The result of a coin toss is either a head or a tail. Here, head and tail are two simple events. These two simple events make up the sample space. Further, the probability of an event occurring falls within the range of 0 to 1. The probability of an...
23.5K
