Notes
  • Artificial-Intelligence
    • Introduction
    • Graph Search
    • Constraint Satisfaction Problems
    • Game Playing
    • Decision Networks and Value of Information
    • Reasoning over Time or Space
    • Naive Bayes
    • Probability and Bayes Nets
  • C++
    • Variables and Basic Types
    • Strings, Vectors, and Arrays
    • Classes
  • Compilers
    • Introduction
    • Lexical Analysis
  • Computer-Architecture
    • RISC-V Instruction Formats
    • Single Cycle RISC-V CPU
    • Instructions and Processors in MIPS
    • Dynamic Scheduling
  • Computer-Networking
    • DNS
    • Email
    • File Transfer Protocol (FTP)
    • Principles of Network Applications
    • The Web and HTTP
    • introduction
    • link-layer-reliable-data-transmission
    • network-layer
    • The Sockets Interface
    • Transportation Layer
  • Data-Structures-and-Algorithms
    • Disjoint Sets
    • Priority Queues and Heaps
    • Sort
  • Database-Systems
    • Introduction
    • Modeling
    • Relational Algebra and SQL
    • Database Administration
    • Transaction
    • Files and Indices
    • Query Processing
    • Query Optimization
    • Normalization
    • Data Warehousing and Dimensional Modeling
    • Distributed Databases
    • NoSQL
  • Distributed-ML
    • Distributed ML
    • PyTorch Distributed Training
  • Introduction-to-Computer-Systems
    • X86-64
    • Exception Control Flow
    • Linking
    • Network Programming
  • LeetCode
    • Dynamic Programming
  • Miscellaneous
    • Glibc Malloc
    • 八股总结
    • Python Memory Management
    • QEMU
  • Operating-Systems
    • Cache
    • Deadlocks
    • File Sytem Concepts
    • Interprocess Communication Cases
    • Inter-process Communication Concepts
    • Interprocess Communication Implementation
    • Exceptions
    • Introduction
    • I/O
    • Case Study: Linux Futex (Fast Userspace Mutex)
    • Case Study: Linux Scheduling
    • Case Study: Linux Task Model
    • Memory Management: Mechanisms
    • Memory Management: Strategies
    • Case Study: Pintos Booting
    • Process and Thread
    • Programming: Process and Thread
    • Singals
    • Scheduling
    • Storage Devices
    • Synchronization and Mutual Exclusion
  • Rust
    • collections
    • Fundamental Types
    • Crates and Modules
    • Enumeration
    • Error Handling
    • Expressions
    • generic-types-and-traits
    • Ownership and Moves
    • References
    • Struct
    • Traits and Generics
  • Swift
    • Swift
  • Papers
    • Access-Pattern-of-Remote-Memory
      • Fast RDMA-based Ordered Key-Value Store using Remote Learned Cache
    • Token-Fusion-for-NER
      • AiluRus: A Scalable ViT Framework for Dense Prediction
      • Adaptive Local-then-Global Token Merging for Efficient Semantic Segmentation with Plain Vision Trans
      • Dynamic Token Pruning in Plain Vision Transformers for Semantic Segmentation
      • EViT: Expediting Vision Transformers via Token Reorganizations
      • Which Tokens to Use? Investigating Token Reduction in Vision Transformers
      • Joint Token Pruning and Squeezing (TPS)
      • Learned Token Pruning for Transformers
      • Paper-List
      • SegViT: Semantic Segmentation with Plain Vision Transformers
      • Token Merging: Your ViT But Faster
Powered by GitBook
On this page
  1. Artificial-Intelligence

Introduction

Agent: an entity that has goals or preferences and tries to perform a series of actions that yeild the best/optimal expected outcome given these goals.

Percept sequence: the complete history of content the agent has perceived.

Environment: the given instantiation of the agent.

Environment type:

  • Fully vs Partially observable

  • Deterministic vs. Stochastic

    The outcome of an action is completely determined by the state of the environment and the action taken by an agent.

  • Episodic vs. Sequential

    Episodic environment can be divided into episodes which have no influence on each other

    Short-term actions do not have long-term consequences

  • Static vs. Dynamic

    Environment doesn’t change while the agent is deliberating

  • Discrete vs. Continuous

    The distinct states of the environmement is finite/infinite

Playing soccer: Partially observable, stochastic, dynamic, sequential, continous

Shopping for used books on the internet: Partially observable, stochastic, dynamic, episodic sequential

World: an environmant and the agents that reside within it create a world.

Simple reflex agent: one that doesn’t think about the consequences of its actions, but rather selects an action based solely on the current state of the world. Example: A basic thermostat used in a home to control the heating or cooling system.

Model-based reflex agent: maintains some internal state which depends on the percept history, useful if the current environment cannot be fully described by the current percept. Example: An autonomous robot vacuum cleaner recording the environment and navigates through the room.

Utility-based agents: compares the desirability of different environment states via a utility function. This allows the comparison of different goal states and action sequences and tradeoffs between different goals. Example: ADC.

Goal-based agents: makes decisions in order to achieve a set of predefined goals, in addition to maintaining internal state.

Problem-solving agents:

Single-state problem formulation: complete observable

A solution is a sequence of actions leading from initial state to goal state.

PreviousArtificial-IntelligenceNextGraph Search

Last updated 9 months ago

A problem is defined by four items: (1) Initial state. (2) Actions(or successor function S(x)S(x)S(x)). (3) Goal test. (4) Path cost (additive).