【论文阅读-REC】<<Recommending music on Spotify with deep learing>>阅读

1、协同过滤

协同过滤不使用item的具体信息，因此可适用性很强，在书籍、电影、音乐上都可用；

协同过滤不适用item的具体信息，因此强者愈强；

冷启动问题无法解决

2、基于内容的推荐

使用声音信号推荐

3、用DL预估偏好

把用户和music各表示成vec

4、提升效率

输入：每个audio，切割成3秒的片段；预测：对这些片段求平均

以梅尔声谱作为输入，横轴是频率，纵轴是时间；

仅在时间维度卷积，不在频率维度卷积，这和图像不一样，图像各个维度内涵一样，音频不一样；

通过下采样，获得时间维度的不变形；

最后一个池化层使用了global tmporal 池化，作者认为音频特征有全局性，不像图片是局部的；

5、训练

Loss：MSE，最小化模型输出向量和CF输出向量的差值

trick：通过时间轴偏移，稍微调整声谱，扩展数据

6、变化（优化尝试）

1）More hidden layers

2）ReLU->maxout unit

3）max-pooling->stocastic pooling

4）拉伸或者压缩时域，扩展数据

5）级联别的CF输出的latent vector

7、分析：我们学到了什么

1）低级表征：最大激活

playlist流派不同，最大激活获取了这种特征（音高）

2）低级表征：平均激活

获取一段时间的平均，获得了和声表征

3）高级表征：

对子音乐类型很有表达能力

8、基于相似度的推荐

效果还不错啦

9、用在哪里

1）推荐集合的来源

2）异常过滤

3）冷启动问题

时间： 2024-09-19 09:26:55

【论文阅读-REC】<<Recommending music on Spotify with deep learing>>阅读的相关文章

【论文阅读-REC】<<DEEP NEURAL NETWORKS FOR YOUTUBE RECOMMENDATIONS>>阅读

1.介绍: YouTube推荐的挑战: scale:很多算法在小数据有用,在youtube无用: freshness:需要对对新上传视频足够敏感: noisy:没有真实的用户反馈:缺少结构化的数据 2.skip 3.候选生成: 之前的模型是基于矩阵分解:YouTube的DL模型的前几层就是使用神经网络模拟这种分解:这可以看成是分解技术的非线性泛化 3.1.把推荐看做多分类: NCE和hs,文字指出hs没有得到nce的效果:YouTube认为,遍历树中不相关节点,使效果变差. 在线预估的时候,并不

论文笔记-Deep Neural Networks for YouTube Recommendations

从各方资料总结了一下大体思路,论文中很多细节还有待细读. 1.引言 youtube视频推荐的三大挑战: (1)规模大:数以亿计 (2)新鲜度:每秒就有很多新视频上传,要考虑用户的实时行为和新视频的推荐,平衡好新视频和好视频.(exploration and exploitation) (3)噪音:用户历史行为很稀疏并且有各种难以观测的隐性因子.而视频本身数据也是非结构化的.推荐算法需要有很好的鲁棒性. 2.系统概览和我们熟知的召回.排序流程是一样的.这里候选集的生成除了使用dnn生成的之外,还

论文阅读--Scalable Object Detection using Deep Neural Networks

Scalable Object Detection using Deep Neural Networks 作者: Dumitru Erhan, Christian Szegedy, Alexander Toshev, and Dragomir Anguelov 引用: Erhan, Dumitru, et al. "Scalable object detection using deep neural networks." Proceedings of the IEEE Confere

Mastering the game of Go with deep neural networks and tree search

Silver, David, et al. "Mastering the game of Go with deep neural networks and tree search." Nature 529.7587 (2016): 484-489. Alphago的论文,主要使用了RL的技术,不知道之前有没有用RL做围棋的. 提出了两个网络,一个是策略网络,一个是价值网络,均是通过自我对战实现. 策略网络: 策略网络就是给定当前棋盘和历史信息,给出下一步每个位置的概率.以前的人似乎是用

On Explainability of Deep Neural Networks

On Explainability of Deep Neural Networks « Learning F# Functional Data Structures and Algorithms is Out! On Explainability of Deep Neural Networks During a discussion yesterday with software architect extraordinaire David Lazar regarding how everyth

（转）Understanding, generalisation, and transfer learning in deep neural networks

Understanding, generalisation, and transfer learning in deep neural networks FEBRUARY 27, 2017 This is the first in a series of posts looking at the 'top 100 awesome deep learning papers.' Deviating from the normal one-paper-per-day format, I'll take

Training Deep Neural Networks

http://handong1587.github.io/deep_learning/2015/10/09/training-dnn.html //转载于 Training Deep Neural Networks Published: 09 Oct 2015 Category: deep_learning Tutorials Popular Training Approaches of DNNs?—?A Quick Overview https://medium.com/@asjad/po

Why are Eight Bits Enough for Deep Neural Networks?

Why are Eight Bits Enough for Deep Neural Networks? Deep learning is a very weird technology. It evolved over decades on a very different track than the mainstream of AI, kept alive by the efforts of a handful of believers. When I started using it a

Introduction to Deep Neural Networks

Introduction to Deep Neural Networks Neural networks are a set of algorithms, modeled loosely after the human brain, that are designed to recognize patterns. They interpret sensory data through a kind of machine perception, labeling or clustering raw

为什么深度神经网络难以训练Why are deep neural networks hard to train?

Imagine you're an engineer who has been asked to design a computer from scratch. One day you're working away in your office, designing logical circuits, setting out AND gates, OR gates, and so on, when your boss walks in with bad news. The customer h