Machine Learning can be defined as an approach to achieve artificial intelligence through systems or software models that can learn from experience to find patterns in a set of data.
- An application of Artificial Intelligence
- An ability to automatically learn and improve from experience.
- Recognizes data patterns and creates rules rather than traditional programming.
- Begins with observations or data such as direct experiences or instructions to look for data patterns.
Google uses artificial intelligence and machine learning in almost all of its applications. Google Photos display photos related to your search terms and animate similar photos from your albums into quick videos. Gmail suggest phrases and complete sentences in emails. Analyze and flag spam messages. Google Assistant can take over real-world tasks such as booking a haircut appointment over phone.
Relationship between Machine Learning and Statistical Analysis
Statistical Analysis collects and scrutinizes the data sample to identify trends and formalization of relationship between variables.
- Is a subset of artificial intelligence in the field of computer science.
- Is associated with high-dimensional data.
- Takes away the deterministic function f out of the equation:
Input (X) —> Output (Y)
- Belongs to the field of mathematics.
- Deals with finding a relationship between variables to predict an outcome.
- Deals with low-dimensional data.
- Tries to estimate the function f:
Dependent Variable (Y) = f (Independent Variable) + Error Function
Process of Machine Learning
- Leverages existing data, images, and videos to train algorithms and models.
- Feeds numerous sets of examples called training sets into the system.
- The Larger the training set, the more accurate the AI Systems.
- Labels each item in a training set as either 0 or 1.
Labeled Training Data consists of Input Data and Expected output. Then this labeled data is processed by Machine Learning Algorithm where algorithm studies the data patterns and works out a logic. Then the Learned Model is generated out of the algorithm which is then used with the test data sets.
In Testing phase, the test data contains all of the inputs and generates output which is system based on the logic derived from the training data. The system classifies the test data based on the patterns learned from the training data. Then the patterns from the test data and logic of learned model to make predictions and derive output.
Types of Machine Learning
- Supervised Learning: – The machine learning program is provided with training data along with the expected output or rules to categorize this data also known as labels. The machine learning systems uses the set of inputs and outputs to predict the output for future unseen inputs.
- Unsupervised Learning: – The machine learning algorithm learns from unlabeled data set, only the input data is used by the algorithm to train the model, the algorithm is expected to find patterns and anomalies from these input data. This method is mostly used in MRI analysis.
- Semi-Supervised Learning: – Is when the training data includes some of the desired output. It is a Hybrid Approach and a combination of supervised and unsupervised learning it uses a combination of labeled and unlabeled data. example of semi-supervised learning is when articles are organized into new topics on the web page.
- Reinforcement Learning: – It is a type of machine learning that allows learning system to observes the environment and learns the ideal behavior. The learning system agent observe environment and selects and takes certain actions and receives rewards in return or penalties in certain scenarios. The feedback is given back to system or agent in a loop. When agent learns the strategy or policy that maximizes rewards overtime and tries to maximize the cumulative reward.