🌻个人主页：相洋同学
🥇学习在于行动、总结和坚持，共勉！
目录
Supervised Learning（监督学习）
Unsupervised Learning（无监督学习 ）
Semi-supervised Learning（半监督学习）
Reinforcement Learning（强化学习）
Fully Connected Neural Network, FCNN（全连接神经网络 ）
Recurrent Neural Network, RNN（循环神经网络）
Convolutional Neural Network, CNN（卷积神经网络）
Generative Adversarial Networks, GANs（生成对抗网络）
Term Frequency-Inverse Document Frequency（TF-IDF）
Word2Vec
Supervised Learning（监督学习）
Supervised Learning is a type of machine learning where the algorithm is trained on a labeled dataset, meaning each training example is paired with an output label. This method enables the model to learn to predict the output from the input data. After training, the model can apply this knowledge to new, unseen data. It's widely used for classification and regression tasks.
- regression -- 回归
Unsupervised Learning（无监督学习 ）
Unsupervised Learning involves training a machine learning model on data without labeled responses. The goal is to allow the model to act on the information without guidance, discovering patterns and relationships in the data by itself. This approach is used in clustering and association tasks where the structure of data is unknown.
- involves -- 涉及
- patterns -- 模式
- approach -- 方法、手段
- clustering -- 聚类
- association tasks -- 关联任务
- structure -- 排列、构造
Semi-supervised Learning（半监督学习）
Semi-supervised Learning sits between supervised and unsupervised learning. It uses a small amount of labeled data alongside a larger volume of unlabeled data for training. This method leverages the strengths of both supervised and unsupervised learning, making it useful for improving learning accuracy with limited labeled data.
- sits between -- 位于...之间
- alongside -- 与...一起
- leverages -- 利用
Reinforcement Learning（强化学习）
Reinforcement Learning is a type of machine learning where an agent learns to make decisions by performing certain actions and receiving rewards or penalties in return. It focuses on finding a balance between exploration of uncharted territory and exploitation of current knowledge. Applications include robotics, gaming, and navigation.
- agent -- 代理
- decisions -- 决策
- actions -- 行动
- rewards -- 奖励
- penalties -- 惩罚
- exploration -- 探索
- exploitation -- 利用
- robotics -- 机器人学
- navigation -- 导航
Fully Connected Neural Network, FCNN（全连接神经网络 ）
Fully Connected Neural Networks (FCNNs) consist of multiple layers of neurons, where each neuron in one layer is connected to all neurons in the next layer. This architecture allows the network to learn complex relationships in the data. FCNNs are widely used for tasks that involve pattern recognition, such as image classification and speech recognition, due to their ability to model complex non-linear relationships.
- consist -- 由什么组成
- multiple layers -- 多层
- architecture -- 架构
- pattern recognition -- 模式识别
- image classification -- 图像分类
- speech recognition -- 语音识别
- non-linear relationships -- 非线性关系
Recurrent Neural Network, RNN（循环神经网络）
Recurrent Neural Networks (RNNs) are designed to recognize patterns in sequences of data, such as text or time series. They achieve this by maintaining a 'memory' of previous inputs using their internal state, which allows them to make predictions about future events in a sequence. RNNs are particularly useful for natural language processing tasks like language modeling and text generation.
- sequences of data -- 数据序列
- text or time series -- 文本或时间序列
- maintaining -- 维持
- 'memory' -- 记忆
- internal state -- 内部状态
- language modeling -- 语言建模
- text generation -- 文本生成
Convolutional Neural Network, CNN（卷积神经网络）
Convolutional Neural Networks (CNNs) are specialized in processing data with a grid-like topology, such as images. CNNs use convolutional layers to filter inputs for useful information without losing the spatial relationship between pixels, making them exceptionally good at tasks like image recognition, image classification, and object detection. Their ability to automatically and adaptively learn spatial hierarchies of features makes them powerful tools in computer vision.
- grid-like topology -- 网格状拓扑
- convolutional layers -- 卷积层
- filter -- 过滤
- spatial relationship -- 空间关系
- pixels -- 像素
- image recognition -- 图像识别
- object detection -- 对象检测
- spatial hierarchies -- 空间层次
- computer vision -- 计算机视觉
Generative Adversarial Networks, GANs（生成对抗网络）
Generative Adversarial Networks (GANs) consist of two neural networks, the generator and the discriminator, which are trained simultaneously through a competitive process. The generator creates data that is similar to, but not identical to, the training data, while the discriminator evaluates the authenticity of the generated data. This setup enables GANs to generate highly realistic images, music, text, or other data types, pushing the boundaries of what's possible in generative models.
- generator -- 生成器
- discriminator -- 判别器
- trained simultaneously -- 同时训练
- competitive process -- 竞争过程
- authenticity -- 真实性
- realistic images -- 真实感图像
- generative models -- 生成模型
- pushing the boundaries -- 推动界限
Term Frequency-Inverse Document Frequency（TF-IDF）
TF-IDF is a statistical measure used to evaluate the importance of a word within a document in a collection or corpus. It increases with the number of times a word appears in the document but is offset by the frequency of the word in the corpus. TF-IDF is widely used in information retrieval and text mining.
- statistical measure -- 统计量度
- importance -- 重要性
- document -- 文档
- collection -- 集合
- corpus -- 语料库
- appears -- 出现
- offset -- 抵消
- information retrieval -- 信息检索
- text mining -- 文本挖掘
Word2Vec
Word2Vec is an algorithm for natural language processing. It converts words into vector space representation, making it possible to analyze words mathematically. This model captures the contextual relationships between words, facilitating tasks such as word prediction, and similarity assessment.
- converts -- 转换
- vector space representation -- 向量空间表示
- analyze -- 分析
- mathematically -- 数学地
- capture -- 捕捉
- contextual relationships -- 上下文关系
- facilitating -- 促进
- similarity assessment -- 相似性评估
以上
君子坐而论道，少年起而行之，共勉
