Here, There are different types of Machine Learning.
Supervised Learning
X, y (pre-classified training examples)
Given an observation x, what is the best label for y?
For every data instance we can have input X and corresponding output y, this is called label data.
In supervised learning you have number of training instance, so this contains the training instances. For training instances each instance comprises of input and output , so this is first training instance to nth training instance.
The given all the training instance the learning algorithm will come up with the model and this model can be use to classify or to find the output or corresponding y value for new observation so, given new input x you can use the model to find out output y, so this is supervised learning.
Unsupervised Learning
- X
- Given a set of x's, cluster or summarize them
We are only given x , there is no label to the data. In unsupervised learning we only have xs different x1, x2, ....xn.this is the data and the learning algorithm produce cluster will group the data.
So best on similarity of the data item to each other we can find out certain group among the data so that is called unsupervised learning.
Reinforcement Learning
- Determine what to do based on rewards and punishments.We have agent which act in the environment, so the agent can take action and this action can impact the environment.
Agent take action and environment goes to new state and give some rewards to the agent.
That reward may be positive reward can be negative reward or can be nothing at particular time state. agent is continually acting in this world.
Reinforcement learner learn a policy that is given a state what action to take? so that not only the short term reward is optimize but the over all utility of the agent over its entire time horizon is optimize.
In reinforcement learning you have an agent who is acting in an environment, and you want to figure out what actions the agent must take at every step and action the agent takes is best on the rewards the agents get in different states.
Semi-supervised Learning
In this learning it is combination of supervised and unsupervised learning that is you have some level training data and you also have larger amount of unlevel training data and you can try to come up with some learning algorithm that can work even when the training data is limited.
In semi-supervised learning we have combination of level data and un-level data.The level data belong two different classes one is circle and another class is triangle.
In semi-supervised learning apart from having data from two classes you also have un-level data which is indicate by the small service. for example supervised learning based on the data supervised data come up some function and if we also have un-level data in addition to the level data we might try to come up with better function.
Supervised Learning
X, y (pre-classified training examples)
Given an observation x, what is the best label for y?
For every data instance we can have input X and corresponding output y, this is called label data.
In supervised learning you have number of training instance, so this contains the training instances. For training instances each instance comprises of input and output , so this is first training instance to nth training instance.
The given all the training instance the learning algorithm will come up with the model and this model can be use to classify or to find the output or corresponding y value for new observation so, given new input x you can use the model to find out output y, so this is supervised learning.
Unsupervised Learning
- X
- Given a set of x's, cluster or summarize them
We are only given x , there is no label to the data. In unsupervised learning we only have xs different x1, x2, ....xn.this is the data and the learning algorithm produce cluster will group the data.
So best on similarity of the data item to each other we can find out certain group among the data so that is called unsupervised learning.
Reinforcement Learning
- Determine what to do based on rewards and punishments.We have agent which act in the environment, so the agent can take action and this action can impact the environment.
Agent take action and environment goes to new state and give some rewards to the agent.
That reward may be positive reward can be negative reward or can be nothing at particular time state. agent is continually acting in this world.
Reinforcement learner learn a policy that is given a state what action to take? so that not only the short term reward is optimize but the over all utility of the agent over its entire time horizon is optimize.
In reinforcement learning you have an agent who is acting in an environment, and you want to figure out what actions the agent must take at every step and action the agent takes is best on the rewards the agents get in different states.
Semi-supervised Learning
In this learning it is combination of supervised and unsupervised learning that is you have some level training data and you also have larger amount of unlevel training data and you can try to come up with some learning algorithm that can work even when the training data is limited.
In semi-supervised learning we have combination of level data and un-level data.The level data belong two different classes one is circle and another class is triangle.
In semi-supervised learning apart from having data from two classes you also have un-level data which is indicate by the small service. for example supervised learning based on the data supervised data come up some function and if we also have un-level data in addition to the level data we might try to come up with better function.
No comments:
Post a Comment