风信子的博客

【论文题目】：Deep learning【论文来源】：Deep learning【翻译人】：BDML@CQUT实验室Deep learningYann LeCun, Yoshua Bengio& Geoffrey Hinton深度学习AbstractDeep learning allows computational models that are composed of multiple processing layers to learn representations of da

论文题目：Machine learning: Trends, perspectives, and prospects论文来源：Machine learning: Trends, perspectives, and prospects【论文翻译】Machine learning: Trends, perspectives, and prospects机器学习：趋势、观点和展望M. I. Jordan * and T. M. Mitchell *AbstractMachine learning ad

论文题目：Machine learning: Trends, perspectives, and prospects翻译人：BDML@CQUT实验室Machine learning: Trends, perspectives, and prospects机器学习：趋势、观点和前景 M. I. Jordan* and T. M. Mitchell*AbstractMachine learning addresses the question of how to build computers th

[论文翻译] A Global Geometric Framework for Nonlinear Dimensionality Reduction论文题目：A Global Geometric Framework for Nonlinear Dimensionality Reduction论文来源：A Global Geometric Framework for Nonlinear Dimensionality Reduction翻译人：BDML@CQUT实验室A Global Geometric

【论文题目】：Clustering by Passing Messages Between Data Points【论文来源】：Clustering by Passing Messages Between Data Points【翻译人】：BDML@CQUT实验室Clustering by Passing Messages Between Data PointsBrendan J. Frey*, Delbert Dueck通过在数据点之间传递消息进行聚类AbstractClustering d

【论文题目】：Clustering by Passing Messages Between Data Points【论文来源】：Clustering by Passing Messages Between Data Points【翻译人】：BDML@CQUT实验室Clustering by Passing Messages Between Data PointsBrendan J. Frey* and Delbert Dueck通过在数据点之间传递消息进行聚类AbstractClusterin

Deep learningYann LeCun, Yoshua Bengio& Geoffrey Hinton 深度学习Yann LeCun, Yoshua Bengio& Geoffrey HintonAbstract  Deep learning allows computational models that are composed of multiple processing layers to learn representations of data with .

Clustering by fast search and find of density peaks基于快速搜索和密度峰发现的聚类方法Alex Rodriguez and Alessandro LaioCluster analysis is aimed at classifying elements in to categories on the basis of their similarity. Its applications range from astronomy to bioinform

论文题目：Clustering by Passing Messages Between Data Points论文来源：Clustering by Passing Messages Between Data Points翻译人：BDML@CQUT实验室Clustering by Passing Messages Between Data PointsBrendan J. Frey* and Delbert Dueck通过在数据点之间传递消息进行聚类Brendan J. Frey* and Del

[论文翻译] Clustering by Passing Messages Between Data Points论文题目：Clustering by Passing Messages Between Data Points论文来源：Clustering by Passing Messages Between Data Points翻译人：BDML@CQUT实验室Clustering by Passing Messages Between Data PointsBrendan J. Frey* a

【论文翻译】：Deep Residual Learning for Image Recognition【论文来源】：Deep Residual Learning for Image Recognition【翻译人】：BDML@CQUT实验室Deep Residual Learning for Image Recognition基于深度残差学习的图像识别2016 IEEE Conference on Computer Vision and Pattern Recognition图像识别的深度残差学

论文题目：Fully Convolutional Networks for Semantic Segmentation论文来源:Fully Convolutional Networks for Semantic Segmentation_2015_CVPR翻译人：BDML@CQUT实验室Fully Convolutional Networks for Semantic Segmentation 用于语义分割的全卷积网络 Jonathan Long∗ Evan Shelhamer∗

Deep Residual Learning for Image RecognitionKaiming He Xiangyu Zhang Shaoqing Ren Jian Sun基于深度残差学习的图像识别Kaiming He Xiangyu Zhang Shaoqing Ren Jian SunAbstractDeeper neural networks are more difficult to train. We present a residual learning framewor

DeepResidualLearningforImageRecognition基于深度残差学习的图像识别AbstractDeeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explici

Clustering by Passing Messages Between Data PointsBrendan J. Frey*, Delbert Dueck通过在数据点之间传递消息进行聚类Abstract: Clustering data by identifying a subset of representative examples is important for processing sensory signals and detecting patterns in data. Suc

[论文翻译] Deep Learning论文题目：Deep Learning论文来源:Deep learning Nature 2015翻译人：BDML@CQUT实验室Deep learningYann LeCun, Yoshua Bengio& Geoffrey Hinton深度学习Yann LeCun, Yoshua Bengio& Geoffrey HintonAbstractDeep learning allows computational models that

DeepSort论文翻译摘要SORT（Simple Online and Realtime Tracking）是一种实用的多目标跟踪方法，它聚焦在简单、有效的算法上。在本中，我们整合了appearance信息来提高SORT算法的性能。因此，我们能够在更长时间遮挡的情况下跟踪目标，并且有效地减少了ID切换的次数。在保持原始架构的思想前提下，我们将大量计算复杂的部分放在了离线预训练阶段，在这个阶段利用大规模人员重识别数据集可以得到一个深度关联度量。在线运行期间，我们在视觉外观（appearance）空间使用

YOLOv4：目标检测的最佳速度和精度摘要随着深度学习的发展，目前已经出现了很多算法（或者训练技巧，tricks）来提升神经网络的准确率。在实际测试中评价一个算法的好坏优劣主要看两点，一是能否在大规模的数据集中起作用（work），二是是否有理论依据。一些算法仅能在某些特定的模型上或者某类特定的问题上运行，亦或是适用于一些小规模的数据集。然而，还有一些算法，例如batch normalization（BN）或者残差连接（residual-connections）已经被用在了不同的模型，任务以及不同的数据集

Fully Convolutional Networks for Semantic Segmentation Jonathan Long∗ Evan Shelhamer∗ Trevor Darrell UC Berkeley {jonlong,shelhamer,trevor}@cs.berkeley.edu Jonathan Long∗

单幅图像基于暗通道先验的去雾Kaiming He, Jian Sun, and Xiaoou Tang, Fellow, IEEE摘要:在本篇论文中,我们提出了一种简单但是有效的图像先验条件——暗通道先验去从一幅输入图像中去雾。暗通道先验是一种对于大量户外有雾图像的统计结果，它最重要的一个观察结果是户外无雾图像的绝大部分区域包含某些像素的亮度值至少在某一个通道上是非常低的。结合这个先验条件与雾天图像模型，我们可以直接估计雾的厚度并且回复一幅高质量的无雾图像。基于各种各样的有雾图像的实验去雾结果证明了所提

论文题目：A Global Geometric Framework for Nonlinear Dimensionality Reduction非线性降维的全局几何框架科学家们在处理大量高维数据时，如全球气候模式、恒星光谱或人类基因分布等，经常会面临维度降低的问题：在高维观测过程中，发现隐藏在其中的有意义的低维结构。人脑在日常感知中也面临同样的问题，从高维感官输入中提取出30,000个听觉神经元或106个视神经纤维，这是数量很少的感知相关特征。在这里，我们描述了一种解决维度降低问题的方法，该方法使用易于

摘要(机器学习解决的问题是如何创造出可以通过经验自动改进的计算机。它是当今发展最快的技术领域之一，位于计算机科学和统计学的交叉点，是人工智能和数据科学的核心。机器学习的最新进展是由新的学习算法和理论的发展以及在线数据和低成本计算的持续爆炸所推动的。数据密集型机器学习方法的采用可以在科学、技术和商业中找到，从而导致在许多行业，包括医疗保健、制造业、教育、金融建模、警察和市场营销等领域进行更多基于证据的决策。)正文机器学习是一门专注于两个相互关联的问题的学科：“如何构建一个通过经验自动改进的计算机系统？”

论文题目：Deep Learning论文来源：翻译人：BDML@CQUT实验室 Deep LeaningYann LeCun1,2, Yoshua Bengio3 & Geoffrey Hinton4,5深度学习Yann LeCun1,2, Yoshua Bengio3 & Geoffrey Hinton4,5Deep learning allows computational models that are composed of multiple processing la

论文题目：Deep Learning论文来源:Deep Learning_2015_Nature翻译人：CQUT 重理工实验室Deep Learning Yann LeCun∗ Yoshua Bengio∗ Geoffrey Hinton 深度学习 Yann LeCun∗ Yoshua Bengio∗ Geoffrey Hinton AbstractDeep learning allows computational models tha

论文题目：Clustering by Passing Messages Between Data PointsClustering by Passing Messages Between Data PointsBrendan J. Frey*, Delbert Dueck通过在数据点之间传递消息进行聚类Brendan J. Frey*, Delbert DueckAbstractClustering data by identifying a subset of representative e

【论文翻译】Deep LearningYann LeCun∗ Yoshua Bengio∗ Geoffrey Hinton深度学习AbstractDeep learning allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. These methods have

Machine learning: Trends,perspectives, and prospectsM. I. Jordan1* and T. M. Mitchell2*机器学习：趋势、观点和前景Abstract  Machine learning addresses the question of how to build computers that improve automatically through experience. It is one of today’s most ra

论文题目：Machine learning: Trends, perspectives, and prospects论文来源：Machine learning: Trends, perspectives, and prospects_2015_Science翻译人：BDML@CQUT实验室[论文翻译] Machine learning:Trends, perspectives, and prospectsMachine learning:Trends,perspectives, and prospe

Clustering by Passing Messages Between Data Points通过传递数据点之间的消息进行聚类Brendan J. Frey* and Delbert DueckAbstract: Clustering data by identifying a subset of representative examples is important for processing sensory signals and detecting patterns in data.