Support Vector Machine Feng Li Shandong University fli@sdu.edu.cn December 28, 2021 Feng Li (SDU) SVM December 28, 2021 1 / 82 Outline 1 SVM: A Primal Form 2 Convex Optimization Review 3 The Lagrange Soft-Margin SVM 6 Sequential Minimal Optimization (SMO) Algorithm Feng Li (SDU) SVM December 28, 2021 2 / 82 Hyperplane Separates a n-dimensional space into two half-spaces Defined by an outward pointing means moving it parallely along ω (b < 0 means in opposite direction) Feng Li (SDU) SVM December 28, 2021 3 / 82 Support Vector Machine A hyperplane based linear classifier defined by ω and b Prediction

Lecture 7: K-Means Feng Li Shandong University fli@sdu.edu.cn December 28, 2021 Feng Li (SDU) K-Means December 28, 2021 1 / 46 Outline 1 Clustering 2 K-Means Method 3 K-Means Optimization Problem Problem 4 Kernel K-Means 5 Hierarchical Clustering Feng Li (SDU) K-Means December 28, 2021 2 / 46 Clustering Usually an unsupervised learning problem Given: N unlabeled examples {x1, · · · , xN}; no. achieves: High within-cluster similarity Low inter-cluster similarity Feng Li (SDU) K-Means December 28, 2021 3 / 46 Similarity can be Subjective Clustering only looks at similarities, no labels are given

Online Serving System � 推荐全链路⾃适应 � 统⼀建模，根据请求量削峰填⾕，资源利⽤最⼤化 [ijcai2021] UNBERT: User-News Matching BERT for News Recommendation [CIKM2021] Self-Supervised Learning on Users’ Spontaneous Behaviors for Multi-Scenario Multi-Scenario Ranking in E-commerce 端上 重排 场景1 场景X [CIKM2021] One Model to Serve All: Star Topology Adaptive Recommender for Multi-Domain CTR Prediction � 预训练模型Bert GPT-3在CV/NLP⼤⾏其道， 相关技术正在进⼊推荐领域 相关技术正在进⼊推荐领域 问题1. 推荐链路的漏⽃ 是对资源的巨⼤浪费 问题2. 结果利⽤ 不充分，响应不 够快 [2021] MC2 -SF: Slow-Fast Learning for Mobile-Cloud Collaborative Recommendation 问题3. ⼏⼗个场 景，独⽴链路 总结 � 千亿级推荐模型应⽤ O1. 千亿级特征（TB级）的模型的在线/离线训练， 在线推理服务和持续上线

Amplification). ‣ CVE-2022-45199 - Pillow before 9.3.0 allows denial of service via SAMPLESPERPIXEL. ‣ CVE-2021-29063 - Mpmath v1.0.0 through v1.2.1 exposes a Regular Expression Denial of Service (ReDOS) vulnerability Amplification). ‣ CVE-2022-45199 - Pillow before 9.3.0 allows denial of service via SAMPLESPERPIXEL. ‣ CVE-2021-29063 - Mpmath v1.0.0 through v1.2.1 exposes a Regular Expression Denial of Service (ReDOS) vulnerability Amplification). ‣ CVE-2022-45199 - Pillow before 9.3.0 allows denial of service via SAMPLESPERPIXEL. ‣ CVE-2021-29063 - Mpmath v1.0.0 through v1.2.1 exposes a Regular Expression Denial of Service (ReDOS) vulnerability

Peter, et al. "How to Train BERT with an Academic Budget." ACL Anthology, Nov. 2021, pp. 10644-52, doi:10.18653/v1/2021.emnlp-main.831. 10 OpenAI GPT-3 API https://openai.com/api/ 9 GitHub Copilot: Wightman, Ross, et al. "ResNet strikes back: An improved training procedure in timm." arXiv, 1 Oct. 2021, doi:10.48550/arXiv.2110.00476. increase to a large positive value, while all other in the denominator Stability-Efficiency Dilemma: Investigating Sequence Length Warmup for Training GPT Models." arXiv, 13 Aug. 2021, doi:10.48550/arXiv.2108.06084. 20 Fast.AI Course: https://github.com/fastai/fastbook/blob/780b7

Survey on Making Deep Learning Models Smaller, Faster, and Better." arXiv preprint arXiv:2106.08962 (2021). It’s time for a hands-on project to apply our recent learnings and measure their impact. We will load_weights(INITIAL_WEIGHTS) baseline500_hist = train(model, tds, vds, epochs=100) Epoch 1/100 2021-11-09 14:44:20.431426: I tensorflow/stream_executor/cuda/cuda_dnn.cc:369] Loaded cuDNN version 8005 load_weights(INITIAL_WEIGHTS) baseline1000_hist = train(model, tds, vds, epochs=100) Epoch 1/100 2021-11-09 15:38:34.694059: I tensorflow/stream_executor/cuda/cuda_dnn.cc:369] Loaded cuDNN version 8005

1 2021年06月 机器学习-机器学习项目流程 黄海广 副教授 2 本章目录 01 机器学习项目流程概述 02 数据清洗 03 特征工程 04 数据建模 3 1.机器学习项目流程概述 01 机器学习项目流程概述 02 数据清洗 03 特征工程 04 数据建模 4 机器学习的一般步骤 5 机器学习的一般步骤 数据搜集 数据清洗

1 2021年07月 机器学习-高等数学回顾 黄海广 副教授 2 高等数学 1.导数定义： 导数和微分的概念 ?′(?0) = lim ??→0 ?(?0+??)−?(?0) ?? （1） 或者：?′(?0) = lim ?→?0 ?(?)−?(?0) ?−?0 （2） 3 高等数学 2.左右导数导数的几何意义和物理意义 函数?(?)在?0处的左、右导数分别定义为：

1 2021年04月 机器学习-KNN算法 黄海广 副教授 2 01 距离度量 02 KNN算法 本章目录 03 KD树划分 04 KD树搜索 3 01 距离度量 02 KNN算法 03 KD树划分 04 KD树搜索 1.距离度量 4 距离度量 欧氏距离(Euclidean distance) ? ?, ? = ෍

..) ○ Most implementations of recent deep learning papers ○ ... References ● Machine Learning 2021 Spring Pytorch Tutorial ● Official Pytorch Tutorials ● https://numpy.org/ Any questions?