Why is game theory important to AI?

 



Introduction

Mathematical game theory is used to simulate how different players will interact strategically in a setting with predetermined rules and consequences.

Different areas of artificial intelligence can benefit from the application of game theory:

  • Multi-agent AI systems.
  • Imitation and Reinforcement Learning.
  • Adversary training in Generative Adversarial Networks (GANs).

In addition, machinelearning models and many situations in daily life can be described using game theory.

A two-person game in which one player challenges the other to locate the best hyper-plane providing him the most tough points to classify can be used to teach a classification technique like SVM (Support Vector Machines). The outcome of the game will then condense into a trade-off between the two players' strategic prowess (eg. how well the fist player was challenging the second one to classify difficult data points and how good was the second player to identify the best decision boundary).

Game Theory

Game Theory can be divided into 5 main types of games:

  • Cooperative vs Non-Cooperative Games: Participants in cooperative games might form alliances to increase their chances of winning (eg. negotiations). Instead of forming alliances, players cannot do so in non-cooperative games (eg. wars).
  • Symmetric vs Asymmetric Games: All players in a symmetric game have identical objectives, hence the winner will be determined only by the strategies they employ to reach those objectives (eg. chess). Instead, in a symmetric games, players have opposing or discordant objectives.
  • Perfect vs Imperfect Information Games: In games with perfect information, everyone can see what the other players are doing (eg. chess). Instead, the actions of other players are concealed in games with imperfect information (eg. card games).
  • Simultaneous vs Sequential Games: The different players can operate simultaneously in simultaneous games. Instead, in sequential games, every player is informed of the previous deeds of every other player (eg. board games).
  • Zero-Sum vs Non-Zero Sum Games: In games with zero sums, if one player earns something, the other players lose. Instead, many players might profit from one another's gains in non-zero sum games.


Comments

Post a Comment

Popular posts from this blog

EXPLAIN HOW DEEP LEARNING WORKS?

Image
  Deep learning networks gain knowledge by identifying complex patterns in the data they process. The networks can develop several degrees of abstraction to describe the data by constructing computational models that are made up of many processing layers. For instance, a convolutional neural network, a type of deep learning model, can be trained using a lot (like, millions) of photos, such as ones with cats. This kind of neural network often picks up information from the pixels in the photographs it collects. It has the ability to categorise sets of pixels that are typical of cat traits, with sets like claws, ears, and eyes indicating the presence of a cat in a picture. The fundamental building block of the brain is a brain cell, often known as a neuron. An artificial neuron or perceptron was created after being inspired by a neuron. Dendrites are employed by biological neurons to receive inputs. A perceptron operates similarly, taking in a variety of inputs, applying a variety of tr
Read more

What is the difference between Strong AI and Weak AI?

Image
The word " Artificial Intelligence " is sometimes misinterpreted, yet the technology is more beneficial than harmful. Although past developments in the field of AI may no longer be relevant, the method has evolved significantly over time. Despite the fact that AI is one of the newest branches of intellectual inquiry its roots may be traced back thousands of years. Strong AI is a speculative form of machine intelligence that argues that machines can develop human consciousness on par with humans. Strong AI refers to computers or programs that have their own mind and can think and complete complicated tasks without the need for human intervention. Strong AI uses complicated algorithms to assist systems operate in a variety of scenarios and strong AI-powered robots can make judgments without the need for human intervention. Artificial intelligence with restricted capability is known as weak AI or narrow AI. Weak AI is the application of advanced algorithms to specific probl
Read more

WHY IS DATA SCIENCE IMPORTANT?

Image
In general, data science can be defined as the process of extracting information and insights from data. It is a combination of several tools, algorithms and machine learning techniques that uncover hidden patterns in raw data. With the introduction of Artificial Intelligence and Deep Learning, the notions of Data Science have reached a pinnacle in recent years. The basic cornerstone of Data Science is the ability to extract significant insights from unstructured and raw data. We will present you with in-depth knowledge about Data Science and its importance in this blog. Because data science can help transform raw data into actionable insights, it has become a must-have for businesses. A Data Scientist is an expert in the fields of statistics and computer science and he employs his analytical skills to address a variety of business challenges. Role of Data Scientist Each firm uses data in a unique way, and the way data is used also differs from one organization to the next.
Read more