Networks
《Zero Stability Well Predicts Performance of Convolutional Neural Networks》
# 《Zero Stability Well Predicts Performance of Convolutional Neural Networks》 ## 文章结构1. 摘要2. 引言3. 预备知识4. 来自现存CNNs的观察5. 零稳定性网络ZeroSNet6. 实验-- 通过零稳定预测性能 ......
HS-GCN Hamming Spatial Graph Convolutional Networks for Recommendation
[TOC] > [Liu H., Wei Y., Yin J. and Nie L. HS-GCN: Hamming spatial graph convolutional networks for recommendation. IEEE TKDE.](https://arxiv.org/pdf/ ......
DeepObfusCode：Source Code Obfuscation Through Sequence-to-Sequence Networks
一、Introduction 代码混淆技术旨在解决代码逆向对抗问题。 本质上，代码混淆技术的目标是：在保持一个程序逻辑结构不变以及完整保存的前提下，同时让攻击者不易识别，以此保护软件的完整性和知识产权。 传统的防护策略包括： 插入空白/冗余的逻辑运算 增加不必要的条件运算等 传统的混淆技术最大的问题 ......
算法_贝叶斯网络学习_bayesian networks
###基本概念 条件概率 联合概率 边缘概率 链式法则 随机变量的独立性 条件独立性 贝叶斯规则、贝叶斯概率推理和贝叶斯网络模型。 stochastic，主要用作形容词，主要意思为“随机的；猜测的” ###R语言包 R语言用lme4多层次（混合效应）广义线性模型（GLM），逻辑回归分析 lme4广义 ......
Exploiting Noise as a Resource for Computation and Learning in Spiking Neural Networks
郑重声明：原文参见标题，如有侵权，请联系作者，将会撤销发布！ https://arxiv.org/abs/2305.16044 Summary Keywords Introduction Results Noisy spiking neural network and noise-driven le ......
【论文阅读】Run, Don't Walk- Chasing Higher FLOPS for Faster Neural Networks1
> # 🚩前言 > > - 🐳博客主页：😚[睡晚不猿序程](https://www.cnblogs.com/whp135/)😚 > - ⌚首发时间： > - ⏰最近更新时间： > - 🙆本文由 **睡晚不猿序程** 原创 > - 🤡作者是蒻蒟本蒟，如果文章里有任何错误或者表述不清，请 t ......
4.4 Graph Nerual Networks(GNN)
# 1. Introduction GNN简单来说就是Graph + Nerual Networks,关键问题就是将图的结构和图中每个节点和边的特征转化为一般的神经网络的输入(张量).  - [备份网址] ......
COMP9444 Neural Networks and Deep Learning
COMP9444 Neural Networks and Deep LearningTerm 2, 2023Project 1 - Characters and Hidden UnitDynamicsDue: Wednesday 5 July, 23:59 pmMarks: 20% of final ......
【五期邹昱夫】CCF-B（IEEE Access'19）Badnets: Evaluating backdooring attacks on deep neural networks
> "Gu, Tianyu, et al. "Badnets: Evaluating backdooring attacks on deep neural networks." IEEE Access 7 (2019): 47230-47244." 本文提出了外包机器学习时选择值得信赖的提供商的重要 ......
【五期邹昱夫】CCF-B（RAID'18）Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks
> "Liu, Kang, Brendan Dolan-Gavitt, and Siddharth Garg. "Fine-pruning: Defending against backdooring attacks on deep neural networks." Research in Att ......
3.1 卷积神经网路 (Convolutional Neural Networks, CNN)
# 1. 概念引入: Image Classification 我们做图像分类时,一般分为三步: * 所有图片都先 rescale 成大小一样 * 把每一个类别表示成一个 one-hot vector(dimension 的长度决定模型可以辨识出多少不同种类的东西) * 将图片输入到模型中 ![im ......
Graph Neural Networks Inspired by Classical Iterative Algorithms
[TOC] > [Yang Y., Liu T., Wang Y., Zhou J., Gan Q., Wei Z., Zhang Z., Huang Z. and Wipf D. Graph neural networks inspired by classical iterative algor ......
Learning to Pre-train Graph Neural Networks 学习如何预训练GNN
 ，通过auxiliary network提高了模型的可塑性。 ## 方法 ### The Formulation of Auxiliary ......
Self-Supervised Hypergraph Convolutional Networks for Session-based Recommendation
[TOC] > [Xia X., Yin H., Yu J., Wang Q., Cui L and Zhang X. Self-supervised hypergraph convolutional networks for session-based recommendation. AAAI, ......
ceph使用ipv6部署，创建osd：unable to find any IPv4 address in networks 'x:x::0/64' interfaces
使用 **ipv6** 部署 **ceph** 多节点集群 ceph.conf 配置文件如下： ``` public_network = 2001::0/64 cluster_network = 2001::0/64 ms_bind_ipv6 = true ``` 在进行osd创建时无报错，创建好之 ......
Global Context Enhanced Graph Neural Networks for Session-based Recommendation
[TOC] > [Wang Z., Wei W., Cong G., Li X., Mao X. and Qiu M. Global context enhanced graph neural networks for session-based recommendation. SIGIR, 202 ......
文献阅读-Inferring Networks of Diffusion and Influence
Authors: Manuel Gomez-Rodriguez, Jure Leskovec, Andreas Krause Authors Info & Claims ACM Transactions on Knowledge Discovery from DataVolume 5Issue 4A ......
Uncovering the Representation of Spiking Neural Networks Trained with Surrogate Gradient
郑重声明：原文参见标题，如有侵权，请联系作者，将会撤销发布！ Published in Transactions on Machine Learning Research (04/2023) ......
[论文阅读] DGFont++ Robust Deformable Generative Networks for Unsupervised Font Generation
## Pre title: DGFont++: Robust Deformable Generative Networks for Unsupervised Font Generation accepted: Arxiv 2022 paper: https://arxiv.org/abs/2212. ......
Efficient Graph Generation with Graph Recurrent Attention Networks
[TOC] > [Liao R., Li Y., Song Y., Wang S., Nash C., Hamilton W. L., Duvenaud D., Urtasun R. and Zemel R. NIPS, 2019.](http://arxiv.org/abs/1910.00760) ......
Combining Label Propagation and Simple Models Out-performs Graph Neural Networks
[TOC] > [Huang Q., He H., Singh A., Lim S. and Benson A. R. Combining label propagation and simple models out-performs graph neural networks. ICLR, 20 ......
《AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural Networks》特征交叉论文阅读
背景 这是一篇利用多头attention机制来做特征交叉的论文 模型结构 AutoInt的模型结构如上图所示，搞模型包含 Embedding Layer、Interacting Layer、Output Layer三个部分，其中Embedding Layer和Output Layer和普通模型没什么 ......
Understanding Structural Vulnerability in Graph Convolutional Networks
Chen L., Li J., Peng Q., Liu Y., Zheng Z. and Yang C. Understanding structural vulnerability in graph convolutional networks. IJCAI, 2021. 概 mean 是在 G ......
Twitter延迟转化论文《Addressing Delayed Feedback for Continuous Training with Neural Networks in CTR prediction》阅读
背景 由于用户的兴趣是实时变化的，现代推荐、广告系统采用了流式更新的方式来捕捉用户实时兴趣的变化。实时训练的方式面临的一个难题就是正样本的回传是有延迟的，一个实时发送的负样本其实是无法确认是否是真的负样本的。也就是说实时观测到的数据流是一个有偏数据流，并不是真实的数据。如果模型在这个有偏分布上学习， ......
[Pix2Pix] Image-to-Image Translation with Conditional Adversarial NetWorks
paper：https://arxiv.org/pdf/1611.07004.pdf [CVPR 2017] code： https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix https://phillipi.github.io/pix2pi ......