![](https://img-blog.csdnimg.cn/20201014180756918.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
论文记录
DeniuHe
加油!
展开
-
【论文复现:Active Learning via Local Structure Recontruction】
Active learning via local structure reconstruction原创 2023-10-29 11:32:18 · 261 阅读 · 0 评论 -
主动学习论文复现(xPAL):Toward optimal probabilistic active learning using a Bayesian approach - 2021 ML
主动学习论文复现原创 2023-10-17 12:15:59 · 224 阅读 · 0 评论 -
随机采样方法(Random Sampling)
原创 2023-05-16 11:13:44 · 301 阅读 · 0 评论 -
Kernel Logistic Regression
核逻辑回归原创 2023-03-21 21:39:01 · 176 阅读 · 0 评论 -
主动学习评价指标 area under the learning curve (ALC)
ALC原创 2023-02-25 12:24:05 · 139 阅读 · 0 评论 -
Expected model change maximization
什么是Expected model change maximization原创 2022-08-16 11:46:43 · 188 阅读 · 0 评论 -
AL中初始outlier会阻止算法收敛
AL中初始outlier会阻止算法收敛原创 2022-07-08 10:10:10 · 102 阅读 · 0 评论 -
AL依赖单一准则往往是不可行的
AL原创 2022-07-07 11:28:26 · 92 阅读 · 0 评论 -
Ordinal rank is closely related to multiclass classification and metric regression
Ordinal rank is closely related to multiclass classification and metric regression原创 2022-06-13 15:39:13 · 140 阅读 · 1 评论 -
asymptotic (infinite-training-sample)
asymptotic (infinite-training-sample)原创 2022-06-13 10:45:19 · 113 阅读 · 0 评论 -
MMC的AL效果真糟糕
原创 2022-05-26 17:05:34 · 99 阅读 · 0 评论 -
关于高斯核函数为正定的定理
原创 2022-04-10 22:01:16 · 734 阅读 · 0 评论 -
exploration and exploitation
原创 2022-04-10 21:25:06 · 274 阅读 · 0 评论 -
矩阵逆增量式更新
原创 2022-04-09 17:12:02 · 104 阅读 · 0 评论 -
评价指标MSE和AUC的参考文献
原创 2022-04-04 19:44:55 · 1225 阅读 · 0 评论 -
Morison-Woodbury formula使用案例
原创 2022-03-03 11:16:10 · 305 阅读 · 0 评论 -
标称属性样本相似性度量
原创 2022-02-28 20:56:29 · 665 阅读 · 0 评论 -
Discriminative Locality Alignment (DLA)
Discriminative Locality Alignment (DLA) reduces to Linear Discriminant Analysis (LDA) under specific parameter setting.原创 2022-02-18 15:24:10 · 309 阅读 · 0 评论 -
Density-weighted uncertainty sampling 文献溯源
Density-weighted uncertainty sampling原创 2022-01-21 12:13:54 · 237 阅读 · 0 评论 -
生成模型收敛于少量的样本?
生成模型收敛于少量的样本?原创 2022-01-19 12:50:25 · 125 阅读 · 0 评论 -
Expected Error Reduction的缺点
Expected Error Reduction的缺点原创 2022-01-16 20:42:47 · 423 阅读 · 0 评论 -
什么是smoothness assumption
smoothness assumption原创 2022-01-16 15:15:25 · 343 阅读 · 0 评论 -
Feature Selection
原创 2022-01-15 20:06:56 · 121 阅读 · 0 评论 -
cosine distance metric 的参考文献
Pang-Ning Tan et al. 2006. Introduction to data mining. Pearson Education India原创 2021-12-12 10:08:48 · 446 阅读 · 0 评论 -
Active Learning with Multiple Views
原创 2021-11-27 20:29:10 · 95 阅读 · 0 评论 -
A KELM Framework for Classification of Hyperspectral Images Using Active learning
# ---------------------------------------------------这是一篇基于多视图的主动学习方法。文章提出了一种 Adaptive Maximum Disagreement Criterion# ----------------------------------------------------自适应最大不一致准则的定义如下:原理:一个样本对应K个视图,每个视图对应一个分类器。在当前迭代中,使用当前K个分类器对各视图上的无标记样本进行预...原创 2021-11-27 20:02:34 · 553 阅读 · 0 评论 -
偶然不确定性(aleatoric uncertainty)和认知不确定性(epistemic uncertainty)
1. Here, we have to distinguish between the aleatoric uncertainty that is caused by high Bayesian error, and the epistemic uncertainty, which is caused by a lack of information[22].We are not able to reduce the aleatoric uncertainty, but we can acquire m原创 2021-11-25 19:20:26 · 3414 阅读 · 0 评论 -
Distance estimation in data sets with missing values
ESD原创 2021-11-04 14:33:11 · 90 阅读 · 0 评论 -
Active leanring 需要一个好的分类器。
1. Active learning algorithm require robust classifier for robust usefulness estimation. Therefore, we choose generative classifiers, namely the Parzen window classifier (PWC)题外As the optimization of the overall performance level is not the scope of th原创 2021-11-03 20:46:25 · 104 阅读 · 0 评论 -
时间复杂度高,计算复杂度高
1. The last strategy belongs to the expected error reduction based methods. The original Value of Information (VoI) criterion as suggested by Joshi et al.[13] selects the instance x_i that minimizes a risk measure defined by them.It has to mentioned that原创 2021-11-03 20:38:25 · 211 阅读 · 0 评论 -
实验结果分析
1. Further,the effectiveness of the proposed approach has been depicted by comparing its performance with state-of-the-art AL algorithms in terms of classification accuracy and computational times as well. The ELM-based AL with different query strategies w原创 2021-11-03 10:39:33 · 176 阅读 · 0 评论 -
算法运行时间的比较描述
原创 2021-11-03 09:52:07 · 338 阅读 · 0 评论 -
计算复杂度分析案例
1. Naively labeling all of the data can be prohibitively expensive.原创 2021-11-02 09:39:21 · 207 阅读 · 0 评论 -
模拟缺失数据MCAR
import numpy as npimport pandas as pdfrom scipy.stats import normfrom scipy.stats import multivariate_normal as mvnfrom scipy.stats import multinomialnp.random.seed(231)K = 3#### For diagnostics. Check if covariance matrix is singular.##def i.原创 2021-10-31 15:47:31 · 414 阅读 · 0 评论 -
Least Confidence-based Sampling 文献
原创 2021-10-31 10:42:11 · 235 阅读 · 0 评论 -
RBF Kernel 是一种度量的证明
原创 2021-10-25 11:11:22 · 637 阅读 · 0 评论 -
Acknowledgements in AI article
1.The authors are grateful to the anonymous reviewers for their comments and suggestions to improve the presentation of this paper.陆续更新原创 2021-10-25 11:07:59 · 104 阅读 · 0 评论 -
What is Random Sampling?
Random sampling的缺点:原创 2021-10-25 10:23:25 · 181 阅读 · 0 评论 -
Active learning setting(English)
1. In many potential applications of machine learning, unlabeled data are abundantly available at low cost, but there is a paucity of labeled data, and labeling unlabeled examples is expensive and / or time-consuming.原创 2021-10-25 09:35:07 · 123 阅读 · 0 评论 -
A two-stage clustering-based cold-start method for active learning
He D, Yu H, Wang G, et al. A two-stage clustering-based cold-start method for active learning[J]. Intelligent Data Analysis, 2021, 25(5): 1169-1185.Abstract:The problem of initialization of active learning is considered in this paper. Especially, this p.原创 2021-09-24 17:02:58 · 187 阅读 · 0 评论