公开(公告)号：US20170039456A1

公开(公告)日：2017-02-09

申请号：US14820972

申请日：2015-08-07

Applicant: Yahoo! Inc.

Inventor： Mohammad Saberian ,  Jia Li ,  Jian Yang

IPC: G06K9/66

CPC classification number: G06N3/08 ,  G06N3/0454 ,  G06N3/084

Abstract: Briefly, embodiments of methods and/or systems of training multiclass convolutional neural networks (CNNs) are disclosed. For one embodiment, as an example, an auxiliary CNN may be utilized to form an ensemble with the collection as a linear combination. The linear combination may be based, at least in part, on boost prediction error encountered during the training process.

Abstract translation: 简单地，公开了训练多级卷积神经网络（CNN）的方法和/或系统的实施例。 对于一个实施例，作为示例，可以使用辅助CNN来形成集合作为线性组合的集合。 线性组合可以至少部分地基于训练过程中遇到的升压预测误差。

公开(公告)号：US20170083752A1

公开(公告)日：2017-03-23

申请号：US14859040

申请日：2015-09-18

Applicant: Yahoo! Inc.

Inventor： Mohammad Saberian ,  Sachin Sudhakar Farfade ,  Jia Li

IPC: G06K9/00 ,  G06T3/40 ,  G06K9/46 ,  G06T7/00 ,  G06K9/66 ,  G06K9/62

CPC classification number: G06T3/40 ,  G06K9/00248 ,  G06K9/4628 ,  G06K9/6272 ,  G06N3/0454

Abstract: Briefly, embodiments of methods and/or systems of detecting and image of a human face in a digital image are disclosed. For one embodiment, as an example, parameters of a neural network may be developed to generate object labels for digital images. The developed parameters may be refined by a neural network to generate signal sample value levels corresponding to probability that a human face may be depicted at a localized region of a digital image.

公开(公告)号：US20170032424A1

公开(公告)日：2017-02-02

申请号：US14815334

申请日：2015-07-31

Applicant: Yahoo! Inc.

Inventor： Suleyman Cetintas ,  Srinath Ravindran ,  Mohammad Saberian ,  Sandeep Soni ,  Kuang-chih Lee ,  Hong Yao ,  Jian Yang ,  Pradhan Pattanayak

IPC: G06Q30/02

CPC classification number: G06Q30/0269 ,  G06Q30/0277

Abstract: The technologies described herein serve contextually relevant advertisements under a guaranteed advertisement campaign. A publisher retrieves a guaranteed advertisement campaign related to a webpage available for serving an advertisement, and identifies a set of advertisements relating to the guaranteed advertisement campaign. Advertisement selecting circuitry of the publisher determines whether an advertisement that is contextually relevant to content published at the webpage is present in the set of advertisements. If there is no contextually relevant advertisement in the set of advertisements, the advertisement selecting circuitry selects an alternative advertisement from the set of advertisements that minimizes an under-delivery risk related to the guaranteed advertisement campaign. If there is a contextually relevant advertisement in the set of advertisements, the advertisement selecting circuitry selects the contextually relevant advertisement. Then, the publisher provides the selected advertisement to a client device.

Abstract translation: 本文描述的技术在保证的广告活动下提供与内容相关的广告。 发布者检索与可用于服务广告的网页有关的保证广告活动，并且识别与保证广告活动相关的一组广告。 发布者的广告选择电路决定在该广告集中是否存在与在网页上发布的内容上下文相关的广告。 如果广告集合中没有上下文相关的广告，则广告选择电路从该广告集合中选择一个替代的广告，使与保证的广告活动相关的不足的风险最小化。 如果广告集合中存在上下文相关的广告，则广告选择电路选择上下文相关的广告。 然后，发布者将所选择的广告提供给客户端设备。