人工智能培训

搜索

论文代码开源:用于点击预测的统一批量在线学习框架(A Unified Batch Online Learning Framework for Click Pre

[复制链接]
admin 发表于 2018-9-15 10:04:49 | 显示全部楼层 |阅读模式
admin 2018-9-15 10:04:49 736 0 显示全部楼层
人工智能论文代码开源:用于点击预测的统一批量在线学习框架(A Unified Batch Online Learning Framework for Click Prediction)请注意该人工智能论文代码开源在github,大部分是python写的,框架可能是tensorflow或者pytorch。最近,已经表明,在超分辨率下,超分辨图像的定量和感知质量之间存在权衡关系,其分别对应于与地面实况图像和自然现象的相似性。在本文中,我们提出了一种新的超分辨率方法,可以提高放大图像的感知质量,同时保留传统的定量性能。该提议的方法使用鉴别器网络和两个定量评分预测器网络,在公司中使用深度网络进行多次通过升级。实验结果表明,该方法实现了定量和感知质量的良好平衡,显示出更为满意的结果。
Recently, it has been shown that in super-resolution, there exists a tradeoffrelationship between the quantitative and perceptual quality of super-resolvedimages, which correspond to the similarity to the ground-truth images and thenaturalness, respectively.In this paper, we propose a novel super-resolutionmethod that can improve the perceptual quality of the upscaled images whilepreserving the conventional quantitative performance.The proposed methodemploys a deep network for multi-pass upscaling in company with a discriminatornetwork and two quantitative score predictor networks.Experimental resultsdemonstrate that the proposed method achieves a good balance of thequantitative and perceptual quality, showing more satisfactory results thanexisting methods.论文代码开源:用于点击预测的统一批量在线学习框架(A Unified Batch Online Learning Framework for Click Prediction) qcWuaO6Td5lUO66a.jpg
URL地址:https://arxiv.org/abs/1809.04789v1     ----pdf下载地址:https://arxiv.org/pdf/1809.04789v1    ----         ----github下载地址:https://github.com/idearibosome/tf-perceptual-eusr    ----    论文代码开源:用于点击预测的统一批量在线学习框架(A Unified Batch Online Learning Framework for Click Prediction)请注意该人工智能论文代码开源在github,大部分是python写的,框架可能是tensorflow或者pytorch,keras,至于具体是哪一个没有完全测试。
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则 返回列表 发新帖

admin当前离线
管理员

查看:736 | 回复:0

快速回复 返回顶部 返回列表