Machine learning types and training
| English | Chinese | Pinyin |
|---|---|---|
| backpropagation | 反向传播 | fǎn xiàng chuán bō |
| supervised learning | 监督学习 | jiān dū xué xí |
| classification | 分类 | fēn lèi |
| unsupervised learning | 无监督学习 | wú jiān dū xué xí |
| clusters | 聚类 | jù lèi |
| reinforcement learning | 强化学习 | qiáng huà xué xí |
| reward | 奖励 | jiǎng lì |
| agent | 智能体 | zhì néng tǐ |
| policy | 策略 | cè lüè |
| gradient descent | 梯度下降 | tī dù xià jiàng |
| loss function | 损失函数 | sǔn shī hán shù |
| epochs | 训练轮次 | xùn liàn lún cì |
How machines learn
- Machine learning comes in three paradigms.
- Each suits a different kind of problem.
- And neural networks learn by backpropagation 反向传播.
AI learning type lab
Classify AI examples by the type of learning or concern involved.
The three paradigms
- Supervised learning 监督学习 — the data has labels (images tagged "cat"/"dog"); the model learns input → label. Used for classification 分类 and regression.
- Unsupervised learning 无监督学习 — no labels; the model finds structure, e.g. clusters 聚类 of similar customers.
- Reinforcement learning 强化学习 — learning by reward 奖励 (below).
Deep learning is part of machine learning, which is part of AI
Supervised learning: a model is trained on labelled data, then recognises new data
Supervised learning uses:
Supervised learning trains on labelled examples (e.g. images tagged cat/dog).
Unsupervised learning is used to:
With no labels, unsupervised learning discovers patterns/clusters in the data.
Match each ML paradigm to its data.
Supervised = labels; unsupervised = no labels; reinforcement = reward feedback.
Reinforcement learning
- An agent 智能体 acts in an environment; each action changes the state and returns a reward.
- The agent learns a policy 策略 (a strategy) that maximises total reward over time — by trial and error, with no labels up front.
- Used for sequential-decision problems: games, robot control, self-driving cars.
A humanoid robot uses AI to see, listen and respond like a person
In reinforcement learning, the agent learns by:
The agent tries actions, receives rewards, and learns a policy that maximises long-term reward.
Training by backpropagation
- Training adjusts the weights so outputs match the targets, using backpropagation with gradient descent 梯度下降:
- forward pass (input → output) → compute the error with a loss function 损失函数 → backward pass (propagate the error back to find each weight's gradient) → update the weights by a small step (the learning rate).
- Repeat over many examples and passes (epochs 训练轮次) until the error stops shrinking. After training, a prediction is just one forward pass.
Industrial robot arms on a production line: reinforcement learning can teach a robot to control its movements
Backpropagation trains a network by:
The error flows from the output back through the network (chain rule) so every weight's gradient is found, then weights step downhill.
In gradient descent the learning rate sets how big a step each weight update takes — too large overshoots the minimum, too small makes training slow.
Backpropagation finds each weight's gradient; the learning rate scales how far down that gradient the weight moves.
You've got it
- supervised = labelled data (classification/regression); unsupervised = find structure (clustering)
- reinforcement = an agent learns a policy by maximising reward through trial and error
- backpropagation: forward pass → loss → propagate error back → update weights (gradient descent)
- training takes many epochs; inference is one forward pass