There is a mistake in the explanation of MAP. 1) The value of .9 was the value of (p(x=0/y=1).p(y=1))p(x=0). [You have shown it to be the value of the conditional probability when y= 0. 2) The 2 probabilities will not add up to 1 since the first expression has p(y=1) in the numerator and the second probability is weighted with p(y=0). 3) Going by the logic, (p(x=0/y=0).p(y=0))p(x=0) = (.25)(.6) / (.2) which works to be .75 Hence, even tho Y will most likely be 1, the work shown by the TA is wrong. That being said, I really do learn a lot from your lectures. Thank you for this :)
Can you please explain why the 2 probabilities will not add up to 1? ( we have the information , that x=0 . now y can only take 2 values , either 0 or 1, so p(y=0|x=0) should be equal to 1 - p(y=1|x=0). What am I missing here? )
There is a mistake in the explanation of MAP.
1) The value of .9 was the value of (p(x=0/y=1).p(y=1))p(x=0). [You have shown it to be the value of the conditional probability when y= 0.
2) The 2 probabilities will not add up to 1 since the first expression has p(y=1) in the numerator and the second probability is weighted with p(y=0).
3) Going by the logic, (p(x=0/y=0).p(y=0))p(x=0) = (.25)(.6) / (.2) which works to be .75
Hence, even tho Y will most likely be 1, the work shown by the TA is wrong.
That being said, I really do learn a lot from your lectures. Thank you for this :)
Can you please explain why the 2 probabilities will not add up to 1?
(
we have the information , that x=0 .
now y can only take 2 values , either 0 or 1,
so p(y=0|x=0) should be equal to 1 - p(y=1|x=0).
What am I missing here?
)
For Y= 1 we have .9 & for Y = 0 we have .75
So Y= 1 ll be the answer
MAP inference is calculated wrongly
Explained better than Mam
Wow! Really!? Means a lot to me! Thanks!
Machine Learning by Prof. Sudeshna Sarkar
Basics
1. Foundations of Machine Learning (ruclips.net/video/BRMS3T11Cdw/видео.html)
2. Different Types of Learning (ruclips.net/video/EWmCkVfPnJ8/видео.html)
3. Hypothesis Space and Inductive Bias (ruclips.net/video/dYMCwxgl3vk/видео.html)
4. Evaluation and Cross-Validation (ruclips.net/video/nYCAH8b5AQ0/видео.html)
5. Linear Regression (ruclips.net/video/8PJ24SrQqy8/видео.html)
6. Introduction to Decision Trees (ruclips.net/video/FuJVLsZYkuE/видео.html)
7. Learning Decision Trees (ruclips.net/video/7SSAA1CE8Ng/видео.html)
8. Overfitting (ruclips.net/video/y6SpA2Wuyt8/видео.html)
9. Python Exercise on Decision Tree and Linear Regression (ruclips.net/video/lIBPIhB02_8/видео.html)
Recommendations and Similarity
10. k-Nearest Neighbours (ruclips.net/video/PNglugooJUQ/видео.html)
11. Feature Selection (ruclips.net/video/KTzXVnRlnw4/видео.html )
12. Feature Extraction (ruclips.net/video/FwbXHY8KCUw/видео.html)
13. Collaborative Filtering (ruclips.net/video/RVJV8VGa1ZY/видео.html)
14. Python Exercise on kNN and PCA (ruclips.net/video/40B8D9OWUf0/видео.html)
Bayes
16. Baiyesian Learning (ruclips.net/video/E3l26bTdtxI/видео.html)
17. Naive Bayes (ruclips.net/video/5WCkrDI7VCs/видео.html)
18. Bayesian Network (ruclips.net/video/480a_2jRdK0/видео.html)
19. Python Exercise on Naive Bayes (ruclips.net/video/XkU09vE56Sg/видео.html)
Logistics Regession and SVM
20. Logistics Regression (ruclips.net/video/CE03E80wbRE/видео.html)
21. Introduction to Support Vector Machine (ruclips.net/video/gidJbK1gXmA/видео.html)
22. The Dual Formation (ruclips.net/video/YOsrYl1JRrc/видео.html)
23. SVM Maximum Margin with Noise (ruclips.net/video/WLhvjpoCPiY/видео.html)
24. Nonlinear SVM and Kernel Function (ruclips.net/video/GcCG0PPV6cg/видео.html)
25. SVM Solution to the Dual Problem (ruclips.net/video/Z0CtYBPR5sA/видео.html)
26. Python Exercise on SVM (ruclips.net/video/w781X47Esj8/видео.html)
Neural Networks
27. Introduction to Neural Networks (ruclips.net/video/zGQjh_JQZ7A/видео.html)
28. Multilayer Neural Network (ruclips.net/video/hxpGzAb-pyc/видео.html)
29. Neural Network and Backpropagation Algorithm (ruclips.net/video/T6WLIbOnkvQ/видео.html)
30. Deep Neural Network (ruclips.net/video/pLPr4nJad4A/видео.html)
31. Python Exercise on Neural Networks (ruclips.net/video/kTbY20xlrbA/видео.html)
Computational Learning Theory
32. Introduction to Computational Learning Theory (ruclips.net/video/8hJ9V9-f2J8/видео.html)
33. Sample Complexity: Finite Hypothesis Space (ruclips.net/video/nm4dYYP-SJs/видео.html)
34. VC Dimension (ruclips.net/video/PVhhLKodQ7c/видео.html)
35. Introduction to Ensembles (ruclips.net/video/nelJ3svz0_o/видео.html)
36. Bagging and Boosting (ruclips.net/video/MRD67WgWonA/видео.html)
Clustering
37. Introduction to Clustering (ruclips.net/video/CwjLMV52tzI/видео.html)
38. Kmeans Clustering (ruclips.net/video/qg_M37WGKG8/видео.html)
39. Agglomerative Clustering (ruclips.net/video/NCsHRMkDRE4/видео.html)
40. Python Exercise on means Clustering (ruclips.net/video/qs7vES46Rq8/видео.html)
Tutorial I (ruclips.net/video/uFydF-g-AJs/видео.html)
Tutorial II (ruclips.net/video/M6HdKRu6Mrc/видео.html )
Tutorial III (ruclips.net/video/Ui3h7xoE-AQ/видео.html)
Tutorial IV (ruclips.net/video/3m7UJKxU-T8/видео.html)
Tutorial VI (ruclips.net/video/b3Vm4zpGcJ4/видео.html)
Solution to Assignment 1 (ruclips.net/video/qqlAeim0rKY/видео.html)
is there no tutorial 5 ?
Sir the most probable inference value of y will be 0
He maked a mistake in map inference
U explain really well!!!! :)
Thankies :)
Great Explanation!!