comp3411考试复习

最新推荐文章于 2024-10-21 15:31:17 发布

LinkEd连接教育

最新推荐文章于 2024-10-21 15:31:17 发布

阅读量785

点赞数

CC 4.0 BY-SA版权

分类专栏：计算机辅导 ai 文章标签： comp3411

本文链接：https://blog.csdn.net/weixin_42089190/article/details/124216244

计算机辅导同时被 2 个专栏收录

14 篇文章

订阅专栏

2 篇文章

订阅专栏

这篇博客汇总了人工智能领域的核心概念，包括代理类型、Prolog编程、搜索算法、约束满足问题、逻辑、规划、决策树、神经网络和语法。重点介绍了不同代理的适用场景，理解Prolog程序的输出，搜索算法的性质，CSP策略的应用，逻辑公式求解，规划表示及算法，强化学习中的Q学习，决策树的构建和修剪，以及神经网络的前向传播和训练规则。此外，还涉及了归纳逻辑编程和各种解析器的运用。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

COMP3411/9814 21T1
Revision and Exam Hints

文章目录

Know your tutorial exercises!

• The tutorial exercises give you the best practice for what questions to
expect in the exam, so …
• Go over your tutorials and make sure you understand the solutions!

Agents

• What are the different types of agents and how do they differ?

Reactive Agent
• Model-Based Agent
• Planning Agent
• Utility-based agent
• Game Playing Agent
• Learning Agent

• When are different types of agents applicable?

Prolog Programming

• Given a program, what is the correct output?
• Given an incomplete program, what additional code is needed to get the
desired output?

Search

• What are the different types of search algorithms?
• What are their properties?
• Order of search
• Optimality
• Complexity
• Can you demonstrate different search algorithms on a simple example?

Constraint Satisfaction

• What are the different CSP strategies?
e.g forward checking, arc consistency
• Can you apply them on a small example?

Logic

• Can you write down the truth table for a formula in propositional logic?
• Do you understand what a model is and can you find it for a given
formula?
• Can you translate a statement in English into logic and vice versa?
• Do you know how to derive a canonical form?
• Can you apply resolution theorem proving to a simple problem?

Planning

• Do you understand how to represent actions (e.g. STRIPS)
• Can you apply a planning algorithm (e.g. forward, backward chaining) to
generate a plan?
Reinforcement Learning
• Can you explain how Q-learning works?
• Can you calculate value functions and Q values?

Decision Trees

• Can you calculate the entropy and information gain to work out how to
split a node in a decision tree?
• Can you calculate errors to decide how to prune a decision tree?

Neural Networks

• Given a network and an activation function, can you propagate value
through a network to get an output (i.e. forward pass)?
• Can you apply the perceptron training rule?
Inductive Logic Programming
• Do you understand inverse substitutions?
• Can you find the least general generalisation of two clauses?
• Can you apply DUCE operators?

Grammars

• Can you work out if a sentence is recognised by a grammar or can the
grammar generate some sentences?
• Can you apply top-down, bottom-up and chart parsers?
• Can you write a simple Definite Clause Grammar?