VIP内容
简介： 机器学习是从数据和经验中学习的算法研究。 它被广泛应用于从医学到广告，从军事到行人的各种应用领域。 CIML是一组入门资料，涵盖了现代机器学习的大多数主要方面（监督学习，无监督学习，大幅度方法，概率建模，学习理论等）。 它的重点是具有严格主干的广泛应用。 一个子集可以用于本科课程； 研究生课程可能涵盖全部材料，然后再覆盖一些。
作者介绍： Hal Daumé III，教授，他曾担任Perotto教授职位，他现在Microsoft Research NYC的机器学习小组中。 研究方向是自然语言处理。
大纲介绍：
- 前言
- 决策树
- Limits of Learning
- 近邻算法
- 感知机
- 联系
- 边缘分类
- 线性模型
- 偏差
- 概率模型
- 神经网络
- 核函数
- 学习理论
- Ensemble 方法
- 高效学习
- 无监督学习
- 期望最大化
- 结构预测
- 模仿学习
- 后记
下载链接： https://pan.baidu.com/s/1QwSGTioJxDCRvlkBqcJr_A
提取码：fwbq
成为VIP会员查看完整内容
热门内容
最新内容
The learned weights of a neural network have often been considered devoid of scrutable internal structure. In this paper, however, we look for structure in the form of clusterability: how well a network can be divided into groups of neurons with strong internal connectivity but weak external connectivity. We find that a trained neural network is typically more clusterable than randomly initialized networks, and often clusterable relative to random networks with the same distribution of weights. We also exhibit novel methods to promote clusterability in neural network training, and find that in multi-layer perceptrons they lead to more clusterable networks with little reduction in accuracy. Understanding and controlling the clusterability of neural networks will hopefully render their inner workings more interpretable to engineers by facilitating partitioning into meaningful clusters.
最新论文
The learned weights of a neural network have often been considered devoid of scrutable internal structure. In this paper, however, we look for structure in the form of clusterability: how well a network can be divided into groups of neurons with strong internal connectivity but weak external connectivity. We find that a trained neural network is typically more clusterable than randomly initialized networks, and often clusterable relative to random networks with the same distribution of weights. We also exhibit novel methods to promote clusterability in neural network training, and find that in multi-layer perceptrons they lead to more clusterable networks with little reduction in accuracy. Understanding and controlling the clusterability of neural networks will hopefully render their inner workings more interpretable to engineers by facilitating partitioning into meaningful clusters.
