Embedded Files
Definition of machine learning
Definition of machine learning
A computer program is said to learn from experience E with respect to a task T and some performance measure P if performance on T, measured by P, improves with experience E
For example, refer here
Machine learning languages
Machine learning languages
Background
Background
Remember derivative meaning
Remember derivative meaning
Matrix basics
Matrix basics
Machine learning algorithms
Machine learning algorithms
Supervised learning
Unsupervised learning
Reinforcement learning
Recommender system
Some questions for strengthening your understanding
Univariate linear regression
Univariate linear regression
Function
Function
y=f(x)
Refer: https://www.youtube.com/watch?v=kHwlB_j7Hkc&list=PLLssT5z_DsK-h9vYZkQkYNWcItqhlRJLN&index=4
Linear regression optimisation
Linear regression optimisation
Gradient descent
Gradient descent
Updation of variables
Updation of variables
After each iteration, find the tangent of the error curve
Simultaneously update all parameters
Understanding gradient
Understanding gradient
If the variable is in left of the local minima, then slope will be negative and so, the variable value will increase and so coming closure to local minima
If the variable is in right of the local minima, then slope will be positive and so, the variable value will decrease and so coming closure to local minima
If the variable will be at local minima, then slope will be zero and so, value will not change ideally
Learning rate
Learning rate
Slow learning rate will converge to local minima, but it will take many iteration
Very high learning rate may not converge or even diverge
Adjustment of learning rate over time
Adjustment of learning rate over time
No need to decrease learning rate since gradient descent will automatically take smaller steps over the time.
Local minima
Local minima
There are many local minima
Based on initial point, solution can converge to different local minima
Multi-variate linear regression
Gradient descent in multivariate case
Gradient descent in multivariate case
Feature scaling
Feature scaling
Why we need to normalise?
Why we need to normalise?
Mean normalisation
Mean normalisation
Gradient descent convergence
Gradient descent convergence
Polynomial regression
Polynomial regression
Normal equation
Normal equation
Gradient descent versus Normal equation
Gradient descent versus Normal equation
Linear regression for classification problem
Linear regression for classification problem
Linear regression is not suitable for classification problem
By luck, below algorithm predicts correctly
Adding a new training data makes prediction wrong
Logistic classification (AKA logistic regression)
Logistic classification (AKA logistic regression)
Hyphothesis value should be in the range of 0 and 1 for decision problems. Logistic regression method ensures this
Non-linear decision boundaries
Non-linear decision boundaries
More complex shapes
More complex shapes
Logistic Regression Cost Function
Logistic Regression Cost Function
Optimisation algorithm
Optimisation algorithm
Multiclass classification
Multiclass classification
Overfitting
Overfitting
Addressing overfitting
Addressing overfitting
Regularised linear regression
Regularised linear regression
Addressing overfit via regularisation
Addressing overfit via regularisation
Solving non-invertibility via regularisation
Solving non-invertibility via regularisation
Motivation for using neural networks
Motivation for using neural networks
With many parameters, the hypothesis function will have lots of features or else hypothesis will not be good fit
Machine needs to covert pixel values to the object
Neural networks
Neural networks
Neural networks processing
Neural networks processing
Reference
Reference
Andrew Ng sessions
https://www.gnu.org/software/octave/
https://www.quora.com/Why-does-Andrew-Ng’s-Machine-Learning-course-use-Octave-instead-of-R
Page updated
Google Sites
Report abuse