相关实验视频
Updated: Jul 2, 2025

13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
9.2K
可扩展的动量传播和对变量分布的分析,用于实际的贝叶斯深度学习
IEEE transactions on neural networks and learning systems
|February 27, 2024
概括
时刻传播 (MP) 提供比蒙特卡洛 (MC) 采样更快的贝叶斯深度学习推断. 本研究引入了扩展批量规范化,用于与MP训练深度模型,实现可比性能和显著降低计算成本.
科学领域:
- 人工智能的人工智能
- 机器学习 机器学习
- 计算统计学 计算统计学
背景情况:
- 贝叶斯深度学习对于量化预测不确定性至关重要.
- 使用蒙特卡洛 (MC) 采样进行变量推断 (VI) 是计算上昂贵的.
- 动量传播 (MP) 提供了一个潜在的更快的替代方案,但在深度模型中面临着挑战.
研究的目的:
- 开发基于MP的快速可靠的贝叶斯深度学习方法.
- 为了应对使用MP训练深度模型的挑战,特别是激活的变异性.
- 调查不同变量分布对MP性能和校准的影响.
主要方法:
- 引入了对随机变量的扩展批量规范化层,以管理深度MP模型中的激活方差.
- 研究了在各种变量分布中对瞬间的处理,以评估预测不确定性质量.
- 对回归和分类任务进行实验,以评估拟议的方法.
主要成果:
- 基于MP的方法在回归任务中实现了与基于MC的方法相当的预测性能.
- 扩展批量规范化使得基于MP的深度模型可以用于分类任务的训练.
- 基于MP的方法比基于MC的方法推断速度快2.0-2.8倍,同时保持预测准确度.
结论:
- 提出的基于MP的贝叶斯深度学习方法是快速可靠的.
- 扩展批量正常化对于训练深度MP模型是有效的.
- 这项工作为可靠性意识的应用程序提供了高效和精确校准的不确定性估计.
相关概念视频
Maxwell-Boltzmann Distribution: Problem Solving
1.5K
Individual molecules in a gas move in random directions, but a gas containing numerous molecules has a predictable distribution of molecular speeds, which is known as the Maxwell-Boltzmann distribution, f(v).
This distribution function f(v) is defined by saying that the expected number N (v1,v2) of particles with speeds between v1 and v2 is given by
This distribution function f(v) is defined by saying that the expected number N (v1,v2) of particles with speeds between v1 and v2 is given by
1.5K
Noncompartmental Analysis: Statistical Moment Theory
105
Noncompartmental analyses leverage statistical moment theory to examine time-related changes in macroscopic events, encapsulating the collective outcomes stemming from the constituent elements in play. Statistical moment theory is a mathematical approach used to describe the time course of drug concentration in the body without assuming a specific compartmental model. SMT provides insights into drug absorption, distribution, metabolism, and elimination by treating drug concentration versus time...
105
Propagation of Uncertainty from Random Error
684
An experiment often consists of more than a single step. In this case, measurements at each step give rise to uncertainty. Because the measurements occur in successive steps, the uncertainty in one step necessarily contributes to that in the subsequent step. As we perform statistical analysis on these types of experiments, we must learn to account for the propagation of uncertainty from one step to the next. The propagation of uncertainty depends on the type of arithmetic operation performed on...
684
Probability Distributions
7.0K
The probability of a random variable x is the likelihood of its occurrence. A probability distribution represents the probabilities of a random variable using a formula, graph, or table. There are two types of probability distribution– discrete probability distribution and continuous probability distribution.
A discrete probability distribution is a probability distribution of discrete random variables. It can be categorized into binomial probability distribution and Poisson...
A discrete probability distribution is a probability distribution of discrete random variables. It can be categorized into binomial probability distribution and Poisson...
7.0K
Propagation of Uncertainty from Systematic Error
519
The atomic mass of an element varies due to the relative ratio of its isotopes. A sample's relative proportion of oxygen isotopes influences its average atomic mass. For instance, if we were to measure the atomic mass of oxygen from a sample, the mass would be a weighted average of the isotopic masses of oxygen in that sample. Since a single sample is not likely to perfectly reflect the true atomic mass of oxygen for all the molecules of oxygen on Earth, the mass we obtain from this...
519
Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
69
Pharmacokinetic models are mathematical constructs that represent and predict the time course of drug concentrations in the body, providing meaningful pharmacokinetic parameters. These models are categorized into compartment, physiological, and distributed parameter models.
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
69

