Artificial Intelligence (AI) is a branch of computer science focused on creating
machines that can simulate human intelligence. It is the science and
engineering of making intelligent computer programs. At its core, AI is
about building systems that can perform tasks that typically require
human intelligence, such as reasoning, discovering meaning, generalizing,
and learning from past experiences.
● Artificial: Made by humans, not occurring naturally.
● Intelligence: The ability to acquire and apply knowledge and skills.
● Simulation: The imitation of a process or situation.
Essentially, AI aims to use computers to understand and replicate human
intelligence.
A Brief History of AI 📜
The journey of AI has been a long one, with several key milestones marking its
progress.
1943: The first recognized work in AI began when Warren McCulloch and Walter
Pitts proposed a model of artificial neurons.
1949: Donald Hebb suggested a way for the connection strength between
neurons to be modified, a concept that became known as Hebbian learning.
1950: Alan Turing introduced the "Turing Test," a test to check a machine's
ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that
of a human.
1956: The term "Artificial Intelligence" was officially coined by John McCarthy at
the Dartmouth Conference, marking the birth of AI as a field.
1966: Joseph Weizenbaum created ELIZA, one of the first chatbots.
1972: Japan built the first intelligent humanoid robot, WABOT-1.
1980: AI made a comeback with "expert systems," which are programs with
decision-making abilities.
1997: IBM's Deep Blue chess computer defeated world champion Garry
Kasparov.
2002: AI entered homes with Roomba, a robotic vacuum cleaner.
2011: IBM's Watson won the quiz show Jeopardy!.
2014: A chatbot named Eugene Goostman won a competition in a Turing test.
Present: AI is now a major part of the business world, with companies like
Google, Facebook, and Amazon heavily investing in deep learning, big data, and
data science to create devices like Alexa and Siri.
AI Agents 🤖
An AI agent is defined as anything that can perceive its environment through
specialized sensors and subsequently act upon that environment through
actuators (or effectors) to achieve predefined goals. Agents are versatile and can
manifest in various forms: a human being, a sophisticated robot, or even a
purely software-based program. The overarching objective of any AI agent is to
,attain high performance and deliver optimized, correct results consistently.
The performance of an AI agent is typically evaluated using a comprehensive
framework known as PEAS:
● Performance: This refers to the criteria used to measure the success of the
agent's actions. It defines what constitutes a desired outcome for the
agent. For example, for a robotic vacuum cleaner, performance might be
measured by the percentage of dirt collected in a given time.
● Environment: This describes the surroundings in which the agent operates.
It includes all the external factors, objects, and conditions that the agent
can perceive and interact with. For instance, the environment for a self-
driving car would include roads, other vehicles, pedestrians, traffic signs,
and weather conditions.
● Actuators: These are the mechanisms through which the agent performs
actions or influences its environment. They are the "output" components
of the agent. Examples include robotic arms, wheels, display screens, or
software commands that change data.
● Sensors: These are the perceptual components that allow the agent to
gather information about its environment. They are the "input"
components. Examples include cameras, microphones, touch sensors,
GPS, or software interfaces that read data from databases.
Types of Agents
There are five main types of agents:
, ● Simple Reflex Agents:
- Description: These agents react purely based on the current input
(percept) without considering any past history of observations. Their
behavior is governed by direct "condition-action" rules.
- How they work: They have a predefined mapping from a specific
perceived condition to a specific action.
- Best suited for: Environments that are fully observable, meaning the
agent can perceive everything relevant to making a decision at any given
moment.
- Example: A thermostat.
- Example Explained: If the thermostat senses the room temperature is
above 25°C (condition), it turns on the air conditioning (action). It doesn't
remember what the temperature was an hour ago or predict future
temperatures; it only reacts to the current reading.
machines that can simulate human intelligence. It is the science and
engineering of making intelligent computer programs. At its core, AI is
about building systems that can perform tasks that typically require
human intelligence, such as reasoning, discovering meaning, generalizing,
and learning from past experiences.
● Artificial: Made by humans, not occurring naturally.
● Intelligence: The ability to acquire and apply knowledge and skills.
● Simulation: The imitation of a process or situation.
Essentially, AI aims to use computers to understand and replicate human
intelligence.
A Brief History of AI 📜
The journey of AI has been a long one, with several key milestones marking its
progress.
1943: The first recognized work in AI began when Warren McCulloch and Walter
Pitts proposed a model of artificial neurons.
1949: Donald Hebb suggested a way for the connection strength between
neurons to be modified, a concept that became known as Hebbian learning.
1950: Alan Turing introduced the "Turing Test," a test to check a machine's
ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that
of a human.
1956: The term "Artificial Intelligence" was officially coined by John McCarthy at
the Dartmouth Conference, marking the birth of AI as a field.
1966: Joseph Weizenbaum created ELIZA, one of the first chatbots.
1972: Japan built the first intelligent humanoid robot, WABOT-1.
1980: AI made a comeback with "expert systems," which are programs with
decision-making abilities.
1997: IBM's Deep Blue chess computer defeated world champion Garry
Kasparov.
2002: AI entered homes with Roomba, a robotic vacuum cleaner.
2011: IBM's Watson won the quiz show Jeopardy!.
2014: A chatbot named Eugene Goostman won a competition in a Turing test.
Present: AI is now a major part of the business world, with companies like
Google, Facebook, and Amazon heavily investing in deep learning, big data, and
data science to create devices like Alexa and Siri.
AI Agents 🤖
An AI agent is defined as anything that can perceive its environment through
specialized sensors and subsequently act upon that environment through
actuators (or effectors) to achieve predefined goals. Agents are versatile and can
manifest in various forms: a human being, a sophisticated robot, or even a
purely software-based program. The overarching objective of any AI agent is to
,attain high performance and deliver optimized, correct results consistently.
The performance of an AI agent is typically evaluated using a comprehensive
framework known as PEAS:
● Performance: This refers to the criteria used to measure the success of the
agent's actions. It defines what constitutes a desired outcome for the
agent. For example, for a robotic vacuum cleaner, performance might be
measured by the percentage of dirt collected in a given time.
● Environment: This describes the surroundings in which the agent operates.
It includes all the external factors, objects, and conditions that the agent
can perceive and interact with. For instance, the environment for a self-
driving car would include roads, other vehicles, pedestrians, traffic signs,
and weather conditions.
● Actuators: These are the mechanisms through which the agent performs
actions or influences its environment. They are the "output" components
of the agent. Examples include robotic arms, wheels, display screens, or
software commands that change data.
● Sensors: These are the perceptual components that allow the agent to
gather information about its environment. They are the "input"
components. Examples include cameras, microphones, touch sensors,
GPS, or software interfaces that read data from databases.
Types of Agents
There are five main types of agents:
, ● Simple Reflex Agents:
- Description: These agents react purely based on the current input
(percept) without considering any past history of observations. Their
behavior is governed by direct "condition-action" rules.
- How they work: They have a predefined mapping from a specific
perceived condition to a specific action.
- Best suited for: Environments that are fully observable, meaning the
agent can perceive everything relevant to making a decision at any given
moment.
- Example: A thermostat.
- Example Explained: If the thermostat senses the room temperature is
above 25°C (condition), it turns on the air conditioning (action). It doesn't
remember what the temperature was an hour ago or predict future
temperatures; it only reacts to the current reading.
