Artificial Intelligence (AI)

Artificial intelligence 人工智能 (AI) builds systems that do tasks normally needing human intelligence — recognising speech and images, translating, playing games, driving, generating text. Most modern AI uses machine learning 机器学习 — algorithms that learn patterns from data instead of being programmed step by step. Within it, deep learning 深度学习, using neural networks 神经网络 with many layers, has been dominant since the 2010s.

A humanoid robot 人形机器人 puts many of these abilities into one body: it uses AI to see faces, understand speech and move its face and arms in a lifelike way.

A humanoid robot uses AI to see, listen and respond like a person

Deep learning is part of machine learning, which is part of AI

Many AI problems sit on a graph 图 — nodes 节点 (states, places) joined by edges 边 (moves, relationships).

• pathfinding: roads form a graph; the shortest route is a graph search (Dijkstra's algorithm, the A* algorithm).
• game playing: each board position is a node, each move an edge; minimax 极小化极大 with alpha-beta pruning searches the game tree.
• state-space search: a planning problem is moving between states by applying operators to reach a goal.
• knowledge representation: a semantic network 语义网络 has concepts as nodes and relationships as edges ("dog IS-A animal"); a knowledge graph 知识图谱 stores facts about the world for search engines and assistants.

AI problems often sit on a graph; here the shortest path is highlighted

Standard tools for navigating graphs include breadth-first search 广度优先搜索 and depth-first search 深度优先搜索.

An ANN is inspired by the brain's neurons. An artificial neuron 人工神经元:

• takes several input values, multiplies each by a weight 权重, and adds them up with a bias term 偏置项.
• applies an activation function 激活函数 (a non-linear function such as ReLU) to the sum.
• outputs the result, which feeds neurons further on.

A single neuron: each input times its weight, summed with a bias, then an activation function

Neurons sit in layers: an input layer, one or more hidden layers 隐藏层 (where useful internal patterns are learned), and an output layer. With many hidden layers it is a deep neural network 深度神经网络, and training it is deep learning.

A neural network with an input layer, two hidden layers and an output layer

ANNs let models learn complex patterns straight from raw data (pixels, audio, text) without hand-designed features — driving breakthroughs in image recognition 图像识别, speech recognition 语音识别, machine translation 机器翻译, and game playing. They do well with large amounts of data, noisy or very complex input, and patterns too hard to capture with explicit rules.

Machine learning

The umbrella term — any algorithm that learns from data. Three paradigms:

• supervised learning 监督学习 — the data has labels 标签 (images tagged "cat"/"dog"); the algorithm learns input → label. Used for classification 分类 (a category) and regression.
• unsupervised learning 无监督学习 — no labels; the algorithm finds structure, e.g. a cluster 聚类 of similar customers.
• reinforcement learning (below).

Use ML when explicit rules would be impractical (spam filters, recommendations, fraud detection).

Supervised learning: a model is trained on labelled data, then recognises new data

Deep learning

A subset of ML using deep neural networks. Lower layers learn simple patterns (edges, phonemes), higher layers combine them into abstract concepts. It needs lots of data and lots of compute (GPUs); for small datasets, simpler ML methods often do better.

Reinforcement learning

In reinforcement learning 强化学习, an agent 智能体 acts in an environment; each action changes the state and returns a reward 奖励. The agent learns a policy 策略 (a strategy) that maximises the total reward over time, by trial and error with no labels up front. Used for sequential-decision problems — games, robot control, autonomous driving.

Reinforcement learning: the agent acts, the environment returns a new state and a reward, and the agent learns from it

A self-driving car 自动驾驶汽车 is a real example. Lidar 激光雷达 and camera sensors (the spinning unit on the roof) build a live picture of the road, and a learned policy decides how to steer, speed up and brake safely.

A self-driving car uses cameras and lidar sensors to see the road around it

Industrial robot arms on a production line: reinforcement learning can teach a robot to control its movements

Training adjusts the weights so outputs match the targets. The standard method is backpropagation 反向传播 (back propagation of errors) with gradient descent 梯度下降. For each training example:

1. forward pass — feed the input through to the output.
2. compute the error with a loss function 损失函数 (a single number for how wrong the output is).
3. backward pass — propagate the error backwards, finding each weight's gradient (how much it contributed to the error) using the chain rule.
4. update the weights by a small step (set by the learning rate 学习率) that reduces the error.

Repeat over many examples and many passes (epochs 训练轮次) until the error stops shrinking. The name "back" comes from step 3: the error flows from the output back towards the input, so every weight's gradient is found in one sweep. After training, a new input needs only one forward pass to get a prediction.

Training adjusts the weights to reach the minimum error

Some tasks predict a number (a house price, tomorrow's temperature) — regression 回归, as opposed to classification (a category).

Linear regression 线性回归 fits a straight line (or hyperplane):

Choose the coefficients to minimise the sum of squared errors against the training data. Use it when the relationship looks roughly linear and you want an interpretable model. For curved data, use polynomial, decision-tree, or neural-network regression methods — same idea: define a model, define a loss, and adjust the parameters to minimise it. Regression and classification are both supervised; the choice depends on whether the answer is a number or a category.

Linear regression fits the line that makes the total squared error (the dashed gaps) as small as possible

Many exam scenarios use the same pattern — a deep-learning model trained on labelled data, often several combined into a pipeline:

• customer identification at an automated shop: the system is trained on labelled face images; a camera captures a face; image recognition extracts a representation; it is matched against registered customers; the closest match identifies the person.
• reading text from images: image recognition finds text regions; optical character recognition 光学字符识别 extracts the characters; machine translation converts them; text-to-speech 文本转语音 reads them aloud.
• checkout item-detection: object-detection AI, trained on labelled product images, sees which items go into a basket and charges the account.

By the time a user interacts with the system, the model is fast — it only does forward-pass inference; the intelligence is in the patterns learned during training.

Worked example. For each task, say whether it needs regression or classification, and what the output layer of an ANN would look like: (a) predict tomorrow's temperature; (b) decide whether an email is spam. Ask what kind of thing is being predicted. (a) A temperature is a number on a continuous scale, so this is regression, and the output layer is a single neuron holding that value. (b) Spam or not-spam is a category, so this is classification, and the output gives a probability per class. Both are supervised learning: each needs labelled examples to train on, and training adjusts the weights by backpropagation to reduce the error. The deciding question is simply number-or-category - not how difficult the task feels.

• Distinguish machine learning, deep learning and reinforcement learning with an example of each.
• Describe an ANN (input, hidden and output layers; weighted connections) and how backpropagation adjusts weights to cut error.
• Distinguish supervised vs unsupervised learning; regression predicts a continuous value.