Machine Learning

Objective of reinforcement learning

. The first is finding previously unknown solutions to existing problems. An example of this learning objective is a machine that plays chess better than any human ever. The second objective of reinforcement learning is finding solutions to problems that arise due to unforeseen circumstances. An example of this learning objective is a machine that is able to find an alternative route through a terrain, after a mudslide has altered the expected route.

6 steps in the machine learning process

1. Data collection 2. Data exploration 3. Data preparation 4. Modeling 5. Evaluation

Machine learning workflow

1. Extract features: choosing features and manipulating dataset. 2. Split dataset: train and test dataset. 3. Train model: Input train dataset into a machine learning model. 4. If desired performance isn't reached, tune the model and repeat step 3.

What is the MNIST

A famous dataset with handwritten digits. It's used by computers to learn to recognize handwritten numbers.

unsupervised data mining

A form of data mining whereby the analysts do not create a model or hypothesis before running the analysis. Instead, they apply the data mining technique to the data and observe the results. With this method, analysts create hypotheses after the analysis to explain the patterns found.

Supervised learning can be broken into 2 parts

Classification and Regression

Unsupervised learning can be broken into 3 parts

Clustering, Association, and Anomaly detection

Exploitation vs exploration

Exploited uses the past details to decide the best decision (maximum reward) and exploration chooses an action with no or less consideration of the reward.

What is 'Overfitting' in Machine learning?

In machine learning, when a statistical model describes random error or noise instead of underlying relationship 'overfitting' occurs. When a model is excessively complex, overfitting is normally observed, because of having too many parameters with respect to the number of training data types. The model exhibits poor performance which has been overfit. Performs great on training data, poor on testing data

What are some ( clustering algorithms? https://machinelearningmastery.com/clustering-algorithms-with-python/

Library Installation, Clustering Dataset, Affinity Propagation, Agglomerative Clustering, BIRCH, DBSCAN, K-Means, Mini-Batch K-Means, Mean Shift, OPTICS, Spectral Clustering, Gaussian Mixture Model

What is dimensionality reduction? When is it used

Reduces the number of features/parameters.

What are the 3 types of machine learning?

Reinforcement learning, supervised learning, unsupervised learning

Objective of supervised learning

Supervised learning is where the objective is to learn patterns in previously labeled data,

What is an SVM

Support Vector Machine. uses classification algorithms for two-group classification problems

What is RMSE? When is it used?

The root-mean-square error (RMSE) is a measure of the differences between values predicted by a model or an estimator and the values actually observed. RMSE is used to evaluate regression models. The lower the RMSE, the better the model. The formula is : rmse = (sqrt(sum(square(predicted_values - actual_values)) / number of observations))

2 parts of Deepmind's mission statement

Understand intelligence, recreate it and use it to solve the worlds problems.

Ensemble methods

Use multiple algorithms to obtain better predictive performance than could be obtained from any of the algorithms by itself

Read this for skills you need in each category

https://towardsdatascience.com/how-to-become-a-machine-learning-engineer-in-2020-1161aa29261e

Objective of unsupervised learning

the objective is to identify unknown patterns in unlabeled data