Model Based Reflex Agent in Artificial Intelligence in HINDI with Real Life Examples

►Artificial Intelligence (Complete Playlist):

Other subject-wise playlist Links:
--------------------------------------------------------------------------------------------------------------------------------------
► Operating System :
►Database Management System:
► Theory of Computation
►Data Structure :
►Computer Networks (Complete Playlist):
►Computer Architecture (Complete Playlist):
►Structured Query Language (SQL):
►Discrete Mathematics:
►Compiler Design:
►Number System:
►Cloud Computing & BIG Data:
►Software Engineering:
►Design and Analysis of algorithms (DAA) (Complete Playlist):
►Graph Theory:
►Programming in C:
►Digital Logic:

---------------------------------------------------------------------------------------------------------------------------------------
Our social media Links:
--------------------------------------------------------------------------------------------------------------------------------------
►For Any Query, Suggestion or notes contribution:

[kuchNayaSekhen]

This guy is seriously unstoppable. Any kind of topic you search gate smashers would be there for offer it's services

[shahryarsaeed]

(00:00) Hello Friends! In this video, we are going to discuss model based reflex agents with real life examples. In the last video, we discussed simple reflex agents. The main point about the simple reflex agent is that it works in a fully observable environment and makes decisions on the basis of what it senses at present and then perform actions.
(00:26) But if we talk about model based reflex agent, this is also a reflex agent. But here we use 'model based'. Model refers to the knowledge base. And how is this knowledge base created? It is created through history. What it has perceived from the environment in the past, it saved it as a knowledge base, as a human brain does.
(00:50) We also create a knowledge base from everything that we have learned in the past. The same knowledge base is considered as the model. It is used for decision-making. But the simple reflex agent works based on what it sees and the if-else condition. But here, these three points are new as compared to the simple reflex.
(01:17) It will perceive things from the environment. But in the simple reflex, this environment was fully observable. In the fully observable environment, the agent has full knowledge of the environment. It knows what's there in the environment. But here's an important point. It's partially observable.
(01:39) Often, a question comes on this that what's the difference between the two and which works in a partially observable environment. In the partially observable environment, the agent doesn't have complete knowledge of the environment. For example, let's discuss the agent. Let's say, there is an agent which is a self-driving car.
(01:57) When a self-driving car drives on the road, it knows everything about its own environment like which sensors are there, there may be high-level sensors and ultrasonic sensors of a high level. It knows about the brakes and acceleration. But, does it know how many cars are there on the road at present? It doesn't.
(02:17) If there is someone in front of it, does it know when he will press the brakes? It doesn't. It can't observe the complete environment. It can only observe certain points. So this point says that in a partially observable environment, model based reflex agent works. And what does it do? It will first perceive or sense something from the environment.
(02:40) Then it will see what the environment is like now. It will analyze the current state of the environment. It will first save that current state. Our mind also first analyzes a situation. After seeing the situation, for a very less time, we store that complete scene. Notice it carefully and visualize in the future that "Yes, you also work like this.
(03:12) " A scenario or situation came. You are storing it in the internal memory. And corresponding to that, you are applying the model base. Here you are using "How the world evolves?" This is about history. You applied the past here. According to that, you have to perform some action. And "What my actions do?" What will your action do, on the basis of that, you will decide what action you need to perform further.
(03:44) "What action I should take now?" Here comes the condition rule. But this condition will not be applied directly. You can't apply the condition on the basis of what you see. a simple example of this is Waymo. Waymo is Google's major project on self-driving cars. But that project is under Waymo and Waymo has already started many cars/cabs in Arizona.
(04:13) They are self-driving. They are driverless cars. What happens in it? What is there in it? If we use the condition action rule, we will give a condition that if there is a car in front of my car, and if it is applying the brakes, the red light behind it will be on and simultaneously, we also have to apply our brakes.
(04:33) This is if-else. Means if the other person is applying brakes, the red light is on, we turned on our red light as well. But, does it happen always? Do you always do it? Or will you instantly apply the brakes completely? This may also be a problem. It may cause an accident. Here, you don't have to act instantly.
(04:50) You have to check the history. You will get to know that it has reduced the speed and you reduced it to 0. No. Similarly, you need to change the lane. You thought that there is a pothole road blockage ahead. You don't have to perform the action instantly. You may also have to decide whether you want to go left or right.
(05:14) If you search for Waymo, you will get the complete video and you will get to know that they have put so many high-level sensors on the car and already implemented it. Those cars are already driving on the road. But, in Arizona, USA, not in India. I don't know when such a project will be passed in India.
(05:38) "Store the percept history." It will store this. So obviously, it needs a storage system too. On the basis of that storage system, it will perform the actions. This is model based in which there are no if-else. You don't need to perform actions instantly. You have to first use your knowledge base.
(05:58) A similar situation came in history, and you have to act on the basis of that. It is also possible that a similar situation may not have occurred in history. What will you do then? You will update your knowledge base. It's not necessary that you have seen everything in your life. But, every person when he sees something new in his life, he stores it in his mind so that it helps him perform actions in a similar situation in the future.
(06:28) This is the main purpose of model-based reflex agent. On the basis of the same, actuators perform the actions. You can say that this is the main difference between simple reflex and model based reflex and how model based works. The Waymo project that I told you about, it is already implemented. And a lot of similar new projects are also being introduced in the market.
(06:53) Thank you!

[sajansekhu]

Sir Aapki vjah se Mera Ai subject favourite bnn gya...Thnkk u so much for wonderful efforts...🙏...Aapki examples se sare concepts clr ho jate h...⭐⭐⭐⭐⭐

[sanghamitraacharya]

Thank you sir because of u I cleared about the basic funda of various agents I really appreciate your lecture videos

[dhulipallasaikiran]

00:01 Model based reflex agents use a knowledge base created through history
00:58 The agent in a partially observable environment lacks full knowledge of the environment.
01:55 Self-driving cars have limited knowledge of their environment
02:39 Perceiving and analyzing the environment and saving its current state.
03:44 Waymo is Google's major project on self-driving cars.
04:33 Applying brakes and changing lanes require careful consideration
05:19 Waymo has implemented high-level sensors on their cars and they are already driving on the road in Arizona, USA.
06:06 Model-based reflex agents store information to help perform actions in the future

[rashigupta]

finally ...I got this topic only because of uu sir ..thank u so muchh

[snobershahzadi]

In a world where we have to pay huge amount of fees to the teachers... U r giving free education dat too quality education... Thnkew so much... Super like

[computerscienceandfuture]

all 18 videos complete I really enjoyed thank you sir

[fuzail]

it looked very funny when u said 'india mein ye project kab implement hoga' lol.

[kpshorts]

mind blowing explanation sir. all the doubts are cleared.... thank you very much

[alifmehraj]

Thank you so much sir. Your videos are very helpful.

[vedantroy]

Thank you sir for great videos. Can you please upload videos on probabilistic reasoning chapter. Thank you once again for your hardwork.

[adithip]

Sir can Knowledge-based agents be similar to model-based agents, If not how are they dissimilar?

[miprincess]

Kamal ki video's hoti ha sir apki

[riturawat]

Sir, I want you to make video on dot net visual basic too☺️

[zkhan]

best one Sir make more videos on Al it's really healpful

[nuwanchathuranga]

Thank you a lot for the explanation brother, Dhanyawad (Thank you)

[PCCOERCoder]

Lecture successfully completed on 11/03/2025 🔥🔥

[chrajagopalrao]

thanks alot bro toc....imp topics 24th net exam humble request am dying to learn toc...just important important say bro so i will see soon

[sarojsharma]

Simple words explanation for complex things