Відео

total waste of time

this is too general, please don't expect much from this. Don't watch if you have exam tomorrow :)

This channel is drastically underrated

Sir aap kamaal ho sir!!!

Its pronounced as Gowsen as in cow.

Excelent video! Thankyou so much!!!!

Excellent video, I learned a lot!

00:01 Unsupervised learning involves learning from unlabeled data 01:27 The video discusses reinforcement learning in Tic-Tac-Toe. 04:06 Understanding the state details and policy in reinforcement learning. 06:47 Importance of subscribing and value of association 08:57 Reinforcement learning involves updating the value of the current state based on the difference between the next and current values. 11:15 Computer learning to play Tic-Tac-Toe 13:19 The video discusses reinforcement learning in Tic-Tac-Toe 15:20 Reinforcement learning has various applications and benefits.

Very well explained. Thanks much :)

Waste of time and data

Does this formula derive from the Bellman's equation in some way? (Q(s,a)=R(s)+lambda×Q(s',a')) Where s and a are state and action corresponding to current state, s' and a' corresponding to the next state and R is the reward

What is the difference between Bayesian Network and Bayesian Belief Network?

thanks

tough concept....yet to catch up completely

bullshit Edited: Im joking

I think you always make the wrong exmple for how an outlier screws up lr. The outlier has to be on the wrong side of the decision boundary. As you showed it, it poses no Problem for lr at All.

Neural network for solving ODE is at very nascent stage. There is a lot to do. Numerical methods serve very well as of now. We don't need neural network for that. Can I have your linkedin ID.

Does anyone have the code for this tic tac toe example?

Very well explained !! Thank you 👍

Why we use ANNs on ODEs. What is difference to get Solution from numerical methode and also get the solution from the ANNs. Why we use ANNs ?

it was effictive

How do we know if the RL algorithm converges on the TicTacToe? Which criteria should we look into?

In the example used, we had the structure of the Bayesian network given and then we could estimate the parameters using BPE. I have read that we can do three things in Bayesian Networks...parameter learning, structure learning and latent structure learning. How do we go about predicting the structure of the Bayesian Network if the parameters are given instead?

Is there a threshold that the generator aspires to achieve in defeating the discriminator? is it possible that such a threshold might be unachievable in certain cases?

Can PCA be used for data compression like K-Means clustering? If it can be used, then which of these two will lead to less loss of data?

simply what does it mean by training a network???plss answer...

Sir, do we always get real eigenvectors of sx? if we get imaginary, a possible workaround exists?

17:32 Temperature is not playing any role in decision making here, does this mean that temperature will also not play any role if the decision is being made in real life by a human? or is this limited to this algorithm.

Is it possible to use K-Means Clustering for classification? Given the number of Clusters, can we assign a subset of clusters into a given class and then maybe use this way for classification, even though this is not the best form of classification we have available.

Can robot be an example of "nearly" strong AI, humans take input via their sense organs, eye, ear, nose, skin and tongue and the data is in form of images, sound, smell, touch/texture and taste. If a robot is able to take data in these format and learn from it and can be rewarded for doing right things and penalised for doing wrong things (for eg: a baby feels pain when he touches fire, therefore learns that he shouldn't touch it), then we can say that robot is an example of strong AI? Do the researchers know what should be the "reward" or "penalty" to a robot?

What is the reason for calling this the "Expectation Maximisation" Algorithm? Which expectation is being maximised?

Sir can you explain a more about Space through model space? I didn't understand.

If all the eigen values are same then out of them, which one would be the principal component?

Does Probabilistic Graphical Model have some connection with Decision Trees as if we can use Probabilistic Graphical Model to find which data set plays an important role in predicting the output same as in Decision Trees where we find Entropy and Information gain. Are these two somewhere related?

Is GANs only used in binary classification or it can also be used in multi class classification? Also won't the computation will increase if we do multi class Classification using GANs.

As stated in post pruning method, the removal of the nodes is done randomly or there is some method to select which node has to be removed. Is there a loss of information.(suppose we remove the 'wind' node from the optimal decision tree stated in the video, does the new decision tree will halt on encountering wind feature)

Can Reinforcement Learning be considered as the ML analogy of Closed-loop control systems (because both take into account the feedback of previous state)?

Is a linear relationship between all variables required for PCA?

Can reinforcement learning be used for feature extraction? And to how much extent is it better than any other approaches?

Does there exist any trade-off between model complexity and dimensionality of the data while using PCA?

The example given in the video contained 3^9 possible states. In real world applications like these, these numbers could go very very high. So instead of going through each and every possibilities which might be very time and energy consuming, is it possible to model our approach in a way which takes into consideration only those samples/states which are much more likely to happen and then proceed to do Re-enforced learning from there?

during training, does the model learn values for every possible state?

Why did we stop at a linear combination of the inputs and not explore combination of higher powers?

How did you get the value of P(W = T) ?

Should we use the primal or the dual form of the svm problem to train a model on a training set with millions of instances and hundreds of features

Is it predetermined to the Discriminator that the Generator always produces fake data (is the Discriminator biased into believing that Generator produces fake data)?

How to know how many times we need to do exploitation and exploration to gain an optimal solution for a particular dataset? Rather than randomly deciding how many times we need to do exploitation and exploration, is there a fixed way to find the which combination or which selection works the best to find the optimal solution.

Sir, If I train an RNN for a task and in it if I add this concept of reward from one state to another, is that considered to be reinforcement learning.

So, are the values of the states determined recursively, after reaching the end state? like if that is the case isn't that similar to training our weights and biases for each node in ANN. But yes, I do understand that here we directly deal with different states.

Is there any way to decrease the correlation by certain design choices of our model Or are these correlations completely determined by the datasets we use?