Machine Learning Note

<Chapter1-Chapter10, 待更> 

Chapter11- 应用机器学习的建议

1） 当机器学习系统工作不如预期的时候怎么做？

• A diagnostic will guide user on which option may work, and which one may not. Then, what to do next? The following.

2) Evaluating a hypothesis: split the data set into: training set (70%), and test set (30%), and see the test error.

3) Model selection problem: how to decide if the degree of  polynomial, how big the lamda, etc.

• Get parameters of each model, test it on the validation set, and choose the one with the lowest validation error.
• Note: It's validation set to evaluate your model, not the test set. Reason: test set would be used to evaluate the generization error (previous section).
• data set -> training/validation/test set (60%/20%/20%). 

 4) Diagnosing bias and variance: know your hypothesis is suffering underfitting or overfitting.

• high bias: training error and validation error are both high.
• high variance: training error is low, but validation error is high.

5) regularizaiton and bias/variance - Choosing lamda

6) learning curve - Error change with sample number increases.

• Good model v.s. High bias v.s. High variance (more training example will help for this case).

7) Revisit - Summary

 NN: Overfitting+Regularization is better than underfitting. The cost is computational expense. 

Chapter12 Machine learning system design

1，The recommended approach to solving machine learning problems is to:

• Start with a simple algorithm, implement it quickly, and test it early on your cross validation data.
• Plot learning curves to decide if more data, more features, etc. are likely to help.
• Manually examine the errors on examples in the cross validation set and try to spot a trend where most of the errors were made.

It is difficult to tell which of the options will be most helpful. A good way is to do the error analysis.

2, Error analysis 

3,  Handling skewed classes

What is skewed classes: one class has a lot of more classes than another, just like the cancer case.

What's its difference: Always output the class with most examples can still get a good accuracy.

A different evaluation error metric: Precision/Recall. 

4, In many application, we want to control how to do the trade off between precision (查准率) and recall (查全率). 

The way to control: adjusting the threshold of the classifying algorithm. 

Is there a way to choose the threshold automatically? Or a similar question is how to compare precision/recall numbers?

• The answer is F1 score. Only compare the one single index.
• Choose the threshold that gives the highest F1 score.

5, Large data's importance: The below graph.

Why is that? - Larege data rationale.

 To better use large data set, two things need to make sure: 

• We train a algorithm with a large number of parameters (to learn complex function).
• The feature contains sufficient information to be able to predict the result correctly.

Chapter13 Clustering - An unsupervised Algorithm 

1, K-means algorithm.

Steps: 1) Cluster assignment; 2) Move centroid.

 2, Optimization objective - Cost function, also called distortion function(distortion of the K-mean algorithm).

Note the meaning of the each step.

What optimization objective can do?

• Knowing if K-mean is working correctly.
• To avoid local optima.

3, Random initialization

• The recommanded way: 
  • 

• 　What can be done if K-mean get stuck by local optima:

  • Try random initialization many times. 

  • Choose the one gives the lowest cost.
  • Note: In the case of small cluster numbers, a few tries (2-5) can always give good results.

4, Choosing the number of the clusters

• Mostly, it is by hand (human insight).
• Sometimes, elbow method will work. The other time, choose the number based on the purpose.

Chapter14 Dimensional Reduction

1, Motivation: Data compression and visualizaiton.

• Data visualization: reduce the dimension form N (N>3) dimensions to k dimensions (k <=3).
• Save memory and speed up the algorithms.

2, Most popular tool or algorithm to do dimensional reduction: PCA (Principle Component Analysis). 

• What is PCA: PCA is to find a lower dimensional surface, onto which to project the data so that the sum of squares of these little blue line segments is minimized.
  • 　　
• Difference between PCA and linear regression.

3, Steps of PCA algorithms

• 

4, Reconstruction from compressed representation

• 

5, How to choose K?

•  

 6, Apply the PCA

• For algorithms speed up, like in ML or linear regression. And save memory.
• Bad use case of PCA: to avoid overfitting.
  • Why? because PCA throws away or reduces the dimension of your data without knowing what the values of y is, it might also throw away some valuable information. The right way is to use regularization to avoid overfitting.
• Do not use PCA untill you have to (use original data first).

 Chapter15 Anomaly Detection

1, What is anomaly detection? - A very common concept used in my work.

• We define a "model" p(x) that tells us the probability the example is not anomalous.
• We also use a threshold ϵ (epsilon) as a dividing line so we can say which examples are anomalous and which are not.

2, Gaussian distribution and anomaly detection algorithms.

• Guassian distribution

• Evaluating an anomaly detection system and determin the threshold (ϵ). 
  • To evaluate our learning algorithm, we take some labeled data, categorized into anomalous and non-anomalous examples.
  • Split the data 60/20/20 training/CV/test and then split the anomalous examples 50/50 between the CV and test sets.

  •  

3, Anomaly Detection vs. supervised learning - how to choose?

•  The cases with a lot of both positive and negative examples - supervised learning.
• The cases with few positivie examples - Anomaly detection.
• 

 4, Choosing what feature to use.

• Check if the features are gaussian distribution: if yes, good, do nothing; if not, try to transform it to let it meet gaussian.
• One common problem is when p(x) is similar for both types of examples (large for both positive and negative cases).
• Error analysis for anomaly detection:
  • Choose features that might take on unusually large or small values in the event of an anomaly.
  • 

5，Multivariate Gaussian Distribution -  Feature's dependency assumption not work.

A very good note: https://www.cnblogs.com/newbyang/p/10338697.html 

<chapter16 skip for now>

 Chapter17 Large Scale Machine Learning

1, Guideline of this chapter:

Why large scale dataset: higher accuracy.

The problem of large scale: computational load.

The things need to do for large scale machine learning: cheaper algorithm, or ways to save time.

• Two ideas: stochastic gradient descent, and map reduce.

 2, Stochastic gradient descent

• Concept：

• A derivative way：Mini-batch gradient descent.
  • Concept: 

  • Can be faster compared with batch GD and stochastic GD:

    • Faster than batch GD: straight forward.

    • Faster than stochastic GD: with appropriate vectorization implementation, mini-batch GD can take the advantage of batch computation.  

3，Stochastic gradient descent converge

4，Online learning 

• 在线学习算法是指，对于一个成型的模型，每次有新的数据来临时，对参数进行更新，更新之后丢弃这个数据，因为不断有新的数据来临，所以不是用整个数据集一次性进行更新参数。
• The algorithm that we apply to it is really very similar to this schotastic gradient descent algorithm, only instead of scanning through a fixed training set, we're instead getting one example from a user, learning from that example, then discarding it and moving on.

 5, Map-Reduce

• Seperate the tasks/data set into multiple machines or machine with multiple cores, to speed up the computation.
• Note: when parallelizing with a multi-core machine, some numerical linear algebra libraries can automatically parallelize their linear algebra operations across multiple cores within the machine.

 Week18 Photo OCR

A good note on Photo OCR: http://www.fanyeong.com/2017/08/08/machine-learning-photo-ocr/

More details:

 1, Pipeline

• A system with several components, several of which are machine learning problem.
• A typical Photo OCR pipeline:
  • text detection -> character segmentation -> character classification.
  • text detection: see the next.
  • character segmentation: supervised learing method. Classifier: can the sliding window split in the middle? 
  • character classification：learned in the previous chapter. The method could be NN, or SVM, etc.

2, text detection method: Supervised learning method.

• try different silding window size to do the sliding window classifier.
  • Resize (re-sample) the patches to a fixed size as the data set before feeding them into the sliding window classifier.
• Mark the region out with text, and expand it to the nearby.
• Filter out the regions with strange aspect ration. Then, done！

3, Artifact data systhesis

• Creating data from scratch.
• Creating data by adding distortions.

4, Ceiling analysis: decide how to spend your resources.