Types of machine learning
Faiz ul Haque Zeya
CEO Transys
The following types of machine learning are
presented in this presentation
• 1.Supervised learning.
• 2.Unsupervised learning.
• 3.Reinforcement learning.
• 4.Semi supervised learning.
• 5.Self supervised learning.
Supervised learning
• Supervised learning algorithms are used when the output is
classified or labeled. These algorithms learn from the past data
that is inputted, called training data, runs its analysis and uses
this analysis to predict future events of any new data within the
known classifications.
• Eg: Image classification: Learn from labelled image and classify
the new image correctly
• Market prediction: Historical data is fed and them learn by the
system and new price for future is predicted.
Unsupervised learning
• The model learn from unlabeled data (we don’t have labeled data)
and the model learn hidden pattern of the data to learn the data.
• Eg:
• Clustering is used to find/create groups of data with similar features
• Generative model. Once distribution of data is known new data can
be generated. This is helpful in missing data.
Reinforcement learning
• The output is compared to find the errors and then is feedbacked to
the system to improve the performance.
Semi-supervised learning
• We have labeled and unlabeled data and lots of unlabeled data and
then algorithms like SELF TRAINING is used to make prediction.
• In self training we have labeled data (small amount) which is used to
make predictions. Then unlabeled data is predicted (pseudo labels
and the process is called pseudo labeling) and correctly predicted
unlabeled data is used to train the model and make further
predictions. There are several iterations of the process.
• The other model is COTRAIN in which two learning
algorithms/classifiers are used. They both uses different features(for
example web page and links) and then pseudo are generated both
and accurate ones are used to update the other.
Self-supervised learning.
• Self-supervised learning is a machine learning process where
the model trains itself to learn one part of the input from another
part of the input. It is also known as predictive or pretext
learning.
• For example, in natural language processing, if we have a few
few words, using self-supervised learning we can complete the
the rest of the sentence. Similarly, in a video, we can predict
predict past or future frames based on available video data.

Types of machine learning.pptx

  • 1.
    Types of machinelearning Faiz ul Haque Zeya CEO Transys
  • 2.
    The following typesof machine learning are presented in this presentation • 1.Supervised learning. • 2.Unsupervised learning. • 3.Reinforcement learning. • 4.Semi supervised learning. • 5.Self supervised learning.
  • 3.
    Supervised learning • Supervisedlearning algorithms are used when the output is classified or labeled. These algorithms learn from the past data that is inputted, called training data, runs its analysis and uses this analysis to predict future events of any new data within the known classifications. • Eg: Image classification: Learn from labelled image and classify the new image correctly • Market prediction: Historical data is fed and them learn by the system and new price for future is predicted.
  • 4.
    Unsupervised learning • Themodel learn from unlabeled data (we don’t have labeled data) and the model learn hidden pattern of the data to learn the data. • Eg: • Clustering is used to find/create groups of data with similar features • Generative model. Once distribution of data is known new data can be generated. This is helpful in missing data.
  • 5.
    Reinforcement learning • Theoutput is compared to find the errors and then is feedbacked to the system to improve the performance.
  • 6.
    Semi-supervised learning • Wehave labeled and unlabeled data and lots of unlabeled data and then algorithms like SELF TRAINING is used to make prediction. • In self training we have labeled data (small amount) which is used to make predictions. Then unlabeled data is predicted (pseudo labels and the process is called pseudo labeling) and correctly predicted unlabeled data is used to train the model and make further predictions. There are several iterations of the process. • The other model is COTRAIN in which two learning algorithms/classifiers are used. They both uses different features(for example web page and links) and then pseudo are generated both and accurate ones are used to update the other.
  • 7.
    Self-supervised learning. • Self-supervisedlearning is a machine learning process where the model trains itself to learn one part of the input from another part of the input. It is also known as predictive or pretext learning. • For example, in natural language processing, if we have a few few words, using self-supervised learning we can complete the the rest of the sentence. Similarly, in a video, we can predict predict past or future frames based on available video data.