By far, it is one of the simplest and most effective explanations I have encountered. You've distilled much key information in a short, easy-to-understand format. Thank you!
@5:29 official subtitles say "high risk given accurate results" while I'm 99% positive the speaker says "higher risk of inaccurate results". Huge divergence of meaning. Wish I knew how to ping the channel directly for clarification/correction.
can you give more details about semi-supervised learning approach please ?? is this HITL ( human in the loop approach) ? whereas a small dataset is labelled where its being used to label and train other larger part of unlabelled dataset ??
In classification, we know the number of available classes a priori (e.g. classifying tumors as either benign or malignant) and when training the model, we specify what class each training example belongs to. On clustering, we may or may not know the number of clusters. Also, when training the algorithms we don't specify what cluster a training example belongs to (since we have no idea). They are similar but not the same.
Great video, very easy to understand! TY.. Any chance you can do a video on "Does Deep Learning suffer from Bias? If so, how? How can we overcome it?" Thanks!!!!
If I make a synthetic copy of the original data (same feature distributions), alter the outcome with random numbers, and work on these artificial data to design an algorithm (i.e., define predictors, combinations of some where components are too rare, and range of values of hyper parameters for example) BEFORE I run the algorithm on the original, labelled data, can this first phase where I am blind to the outcome be named unsupervised learning?
wouldn't it be supervised learning as well? Since, it is labelled irrespective of how accurate your training data is, machine is going to learn from it.
but why unsupervised data could not detect regression? If it shows out of average value, then that data might be a regression point. I am new to this and tried to use elk on my job, thanks
I had the same question and found this: ruclips.net/video/8bGNI_w36n0/видео.htmlsi=McTXHfWFHk8m2jb9&t=120 which shows they write forward and then mirror the video.
![BLACK BAG - Official Trailer [HD] - Only in Theaters March 14](http://i.ytimg.com/vi/Du0Xp8WX_7I/mqdefault.jpg)
I love the way the content is delivered, very logical and clear. Thank you very much!
This is one of the best video on machine learning, short and precise 👍
I wish I can like this more than once
By far, it is one of the simplest and most effective explanations I have encountered. You've distilled much key information in a short, easy-to-understand format. Thank you!
This instructor in particular has a wonderful way of explaining the topics very clearly and plainly.
IBM's content are the best I've seen
Simple explanation, just what I needed.
Simple, precise and to the point explanation, I really love these videos by IBM, thank you.
How is he writing in reverse?? I like his way of explanation!!!!
Much appreciated for such a useful tutorial video.
Thanks IBM.
This is an amazing explanation
Thank you for your explanation. Easier to understand than my uni coursebook.
Appreciate the clear and concise explanation! Your enthusiasm made the topic come alive, You're a great teacher thanks for sharing your expertise.
It is way better than the university teachers .
Thank you Martin! Your explanation is great
Really helpful and quick. Thanks for the explanation!
thanks for the wonderful subtitles. It helps me a lot to study this topic in English.
U have the great skill to explain the subjects. Tnk u.
How can he write in a speculare way so smoothly?
Search on "lightboard videos".
Machine Learning
Maybe he wrote regularily and they mirrored the video.
What does Speculate even mean ??? I don't know much english
@@scooploopslike backwards
So well explained, thank you!
@5:29 official subtitles say "high risk given accurate results" while I'm 99% positive the speaker says "higher risk of inaccurate results". Huge divergence of meaning. Wish I knew how to ping the channel directly for clarification/correction.
very nice video. great explanation and cool visual illustration
Very well explained👍
Nice clear description. Thank you!
You said Logistic Regression is for regression task but actually logistic Regression is a linear model for binary classification task .Thank you
Quite rightly so. Logistic regression outputs in binary which logically is a category/classification of true and false
Correct, logistic regression actually a classification model.
Amazing explanation!! thank you
Thank you, thank you, thank you for the clear explanations.
thank you very much. It was very useful, helpful, clear and quick
I learn from you
Surprisingly 😊
This is really an amazing video! Thank you so much! :-) really grateful for your help
Very well explained
Thank you!
can you give more details about semi-supervised learning approach please ?? is this HITL ( human in the loop approach) ? whereas a small dataset is labelled where its being used to label and train other larger part of unlabelled dataset ??
very useful and short to learn thankyou
great video and great teacher
Always amazing explanations
Watching these whilst I have my exam tommorow😊
hw was ur exam?
Excellent explanation!
Superb explaination!
Watching this whilst I have my exam in 10 min
wow..very impressive...thank you
it is very helpful.Thank you
Loved this! Thank you!
very good content, thank you
classification sounds similar to clustering, what is the difference tho?
In classification, we know the number of available classes a priori (e.g. classifying tumors as either benign or malignant) and when training the model, we specify what class each training example belongs to.
On clustering, we may or may not know the number of clusters. Also, when training the algorithms we don't specify what cluster a training example belongs to (since we have no idea).
They are similar but not the same.
How can I do Ph.D under Martin keen ,Master Inventor?
thanks for explanation it is clear
Thank you ❤️✋
Thanks a lot sir 💌
Thank you
Well, can't you make predictions with association rules ?
Great video, very easy to understand! TY..
Any chance you can do a video on "Does Deep Learning suffer from Bias? If so, how? How can we overcome it?"
Thanks!!!!
If I make a synthetic copy of the original data (same feature distributions), alter the outcome with random numbers, and work on these artificial data to design an algorithm (i.e., define predictors, combinations of some where components are too rare, and range of values of hyper parameters for example) BEFORE I run the algorithm on the original, labelled data, can this first phase where I am blind to the outcome be named unsupervised learning?
wouldn't it be supervised learning as well? Since, it is labelled irrespective of how accurate your training data is, machine is going to learn from it.
I think this could also qualify to be a semi supervised scenario
good explaination
Thanks !👍🙏
but why unsupervised data could not detect regression? If it shows out of average value, then that data might be a regression point. I am new to this and tried to use elk on my job, thanks
Regression in this context means predicting a continuous (non-discrete) value. It's not exactly the regression known is statistics.
thank you brother!
I want to do climate change analysis what should I use ?
I want to know ho does this delivery of the video works …. I mean does he know to type mirrored or what is it … confused a lot
See ibm.biz/write-backwards
Thank you so much man …. Finally got answer to this 😅😅
I always wondered how it worked
Quick response tho 💯💯
amazing. thank you.
Thanks
Great ! Thank u.
Supervised = your dataset has a target variable
Unsupervised = your dataset does not have a target variable
Thx a lot man
5:00 you should have a video explaining what unlabeled data is. Here's a video I found helpful ruclips.net/video/vSO8dFTtlfE/видео.html
Awesome
yooo how are you able to write like that??
I had the same question and found this: ruclips.net/video/8bGNI_w36n0/видео.htmlsi=McTXHfWFHk8m2jb9&t=120 which shows they write forward and then mirror the video.
👍
still feeling a bit slow in 1.5x
The vid is literally 7 minutes
I'm a bit distracted by his ability to write backwards so easily and quickly.
de-click your audio plz, i can hear your saliva
I love the way the content is delivered, very logical and clear. Thank you very much!