Machine learning
Download as PPTX, PDF•0 likes•38 views
The document provides an overview of machine learning, defining it as the study that enables computers to learn from data without explicit programming. It outlines different types of machine learning, including supervised, unsupervised, semi-supervised, and reinforcement learning, along with examples for each type. The process of machine learning, from data acquisition to model training and testing, is also summarized.
Machine learning
- 2. ▪ “The field of study that gives computers the ability to learn without being explicitly programmed” – Arthur Samuel ▪ “A computer program is said to learn from experience ‘E’ with respect to some class of tasks ‘T’ and performance measure ‘P’, if its performance at tasks in T, as measured by P, improves with experience E.” – Tom M. Mitchell What is Machine Learning Let’s try to understand the definition given by Tom Mitchell with couple of examples…
- 3. Example-1: Classifying Emails as Spam or Not Spam: ▪ Task (T): Classifying emails as spam or not spam ▪ Experience (E): Watching the user to mark/label the emails as spam or not spam ▪ Performance (P): The number or fraction of emails correctly classified as Spam or Not Spam Example-2: Recognizing Hand-written Digits : ▪ Task (T): Recognizing Hand written digits ▪ Experience (E): Watching the user to mark/label the hand-written digits to 10 classes (0-9) & identify the underlying pattern ▪ Performance (P): The number or fraction of hand-written digits correctly classified
- 4. When do we use Machine Learning Machine learning is used in cases where: ▪ There is an intuition that a certain rule exists ▪ But, we do not know it or cannot express it mathematically So, the machine / computer learns the rule from data… The performance of the learning agent (in this case the machine / computer) should improve based on experience in either of the following ways: 1. The range of the performance is expanded: The machine can do more. 2. The accuracy on tasks is improved: The machine can do things better. 3. The speed is improved: The machine can do things faster.
- 5. Different kinds of Learning Machine Learning is a “data oriented” discipline. Depending upon how the machine learning models learn the rule inherent in the dataset there are various types of learning methods: ▪ Supervised Learning: In this type of learning the ground truth of the training dataset is known. i.e. the dataset is already labelled. ▪ Un-supervised Learning: In this type of learning the ground truth of the training dataset is not known. i.e. the dataset is not labelled. ▪ Semi-supervised Learning: In this type of learning few of the training dataset are labelled. ▪ Reinforcement Learning: In this type of learning the agent will be penalised/rewarded based on the incorrect/correct prediction. The task of the agent is to maximize the reward.
- 6. Supervised Learning: Learning under supervision Teacher teaches child Child recognizes an apple when she sees it again The machine learning model is also able to learn from past data and make prediction / classification. This is called Supervised Learning.
- 7. Examples of Supervised Machine Learning Sl. No. Age Weight (kg) Systolic Blood Pressure (mmHg) 1 39 67 144 2 47 78 180 3 45 61 138 . . . . . . . . . . . . 240 21 72 148 241 44 63 127 242 63 72 170 ▪ Task: Predicting the blood pressure of an adult based on his/her Age and Weight. ▪ Here Age and Weight are two predictors / input variables. ▪ Here Blood Pressure is the target / output variable. ▪ There are total 242 numbers of examples present in the dataset. ▪ Each training sample is of the form: ⟨(Age, Weight), Blood Pressure)⟩ ▪ In the supervised learning framework we shall use a large fraction of our data to “teach” the model, so that the model can learn and predict the rest of the data with high confidence.
- 8. Examples of Supervised Machine Learning Sl. No. Age Monthly Income (Rs.) Existing Loan (Lakh) Loan Sanction (1 = Yes, 0 = No) 1 39 47000 2.5L 0 2 47 32000 0L 0 3 45 65000 3L 1 . . . . . . . . . . . . . . . 998 21 12000 0L 0 999 44 38000 1L 1 1000 63 28000 0L 0 ▪ Task: Whether to sanction Loan or not to a person based on his/her Age, Monthly Income and Existing Loan amount. ▪ Here the predictor variables are Age, Monthly Income and Existing Loan and the target variable is Loan Sanction. ▪ There are total 1000 labelled samples given. ▪ Each sample is of the form: ⟨(Age, Monthly Income, Existing Loan), Loan Sanction⟩ ▪ Like previous, here also we take a fraction of the total dataset as training samples to train our supervised learning model.
- 9. Supervised Machine Learning ▪ What is the main difference between the examples of Supervised Learning we just saw? ▪ In the first example of “Blood pressure prediction”, the predicted or output variable assumes continuous values within a given range. ▪ While in the second example of “whether to sanction loan or not”, the predicted or output variable is categorical in nature, as it assumes only finite discrete values. This type of supervised learning is known as Prediction or Regression problem This type of supervised learning is known as Classification problem
- 10. Unsupervised Machine Learning: Learning without teacher (self-organization) Unsupervised learning is the machine learning task of inferring a function to describe inherent hidden structure from unlabeled data. Make Two Groups Group- Red Group- Green Group- Apple Group- Tomato
- 11. Unsupervised Machine Learning: Natural grouping of objects into two or more groups such that each object fall into exactly one group is called clustering. The word natural is not quite definitive. It can have different meanings. Hence different types of clusters possible out of same dataset. Apart from clustering there are other unsupervised learning techniques. One of such is Dimensionality Reduction. In dimensionality reduction we try to find out a lower order representation of the data without loosing much information.
- 12. Machine Learning: Learning from data Supervised Learning: Learning with teacher Unsupervised Learning: Learning without teacher Prediction/ Regression: Continuous valued output Classification: Discrete valued output Clustering: Natural grouping of the data Dimensionality Reduction: Finding a lower order representation of the data
- 13. Machine Learning Process ▪ Machine Learning is a data oriented discipline. So we need data to start a Machine Learning process. Usually more data leads to better algorithm. ▪ First identify the task and acquire the required data for the task. ▪ If it’s a supervised learning task then, given the pre-processed data first break it into train and test datasets: (Like 70% for training and 30% for test) ▪ Then pre-process and clean the data. ▪ Train your model on the training dataset and test it on the test dataset. ▪ Keep changing your model / model parameters until you get desired accuracy on both training and test sets.
- 14. Machine Learning Process Training Test Train the Model Validate the Model Done yes no Select a Model Given Data