Machine Learning

Machine learning is a sub-filed of Artificial Intelligence. It's a field of inquiry that's aimed at understanding and building methods that learn how to leverage data to improve performance on a set of tasks. It is the use and development of computer systems that are able to learn and adapt without following explicit instructions, by using algorithms and statistical models to analyze and draw inferences from patterns in data. It was defined in the 1950s by AI pioneer "Arthur Samuel" as “the field of study that gives computers the ability to learn without explicitly being programmed.”

Chatbots, predictive text, language translation apps, the show suggestions by most OTT platforms based on individual user preferences, our social media feeds are all possible because of Machine Learning. It also powers autonomous vehicles and machines that can diagnose medical conditions based on images.

Most companies that deploy Artificial Intelligence (AI) programs are using Machine Learning (ML). Machine learning is a subfield of AI that gives computers the ability to learn without explicitly being programmed. In the last 5-10 years, ML has become a critical and most important way, most parts of AI are done and hence people tend to use AI and ML in a synonymous way, often used interchangeably. Machine learning methods are beginning to be used for various aspects of survey research including responsive/adaptive designs, data processing and nonresponse adjustments and weighting.

Seven Steps of Machine Learning

Machine learning starts with data — numbers, photos, or text, like bank transactions, pictures of people or even bakery items, repair records, time series data from sensors, or sales reports. The data is gathered and prepared to be used as training data, or the information the machine learning model will be trained on. The more data, the better the program. From here, programmers choose a machine learning model to use, supply the data, and let the computer model train itself to find patterns or make predictions. Over time the human programmer can modify the model (if needed), including changing its parameters, to help push it toward more accurate results.

1. Data Collection ---> The quantity & quality of your data dictate the accuracy of your model. ---> The outcome is generally a representation of data that will be used for training. --- > Using pre-collected data, by way of datasets from Kaggle, UCI, etc.,

2. Data Preparation --- > Collect data and prepare it for training. --- > Clean data - remove duplicates, correct errors, deal with missing values, normalization, data type conversions, etc. --- > Randomize data to erase the effects of the particular order in which the data was collected/prepared. ---- >Visualize data

3. Choose a Model --- > Different algorithms suit different tasks; choose the right one.

4. Train the Model --- > The goal of training is to answer a question or make a prediction correctly as often as possible. --- > Each iteration of process is a training step.

5. Evaluate the Model --- > Uses some metric or combination of metrics to "measure" objective performance of model. --- > Test the model against previously unseen data.

6. Parameter Tuning --- > This step refers to hyperparameter tuning. --- > Tune model parameters for improved performance. --- > Simple model hyperparameters include number of training steps, learning rate, initialization values and distribution, etc.

7. Make Predictions --- > Test set data are used to test the model for a better approximation of how the model will perform in the real world.

There are three subcategories of machine learning:

1. Supervised machine learning - models are trained with labeled data sets, which allow the models to learn and grow more accurate over time. Supervised machine learning is the most common type used today.

2. Unsupervised machine learning - a program looks for patterns in unlabeled data. Unsupervised machine learning can find patterns or trends that people aren’t explicitly looking for.

3. Reinforcement machine learning - trains machines through trial and error to take the best action by establishing a reward system. Reinforcement learning can train models to play games or train autonomous vehicles to drive by telling the machine when it made the right decisions, which helps it learn over time what actions it should take.

ML is best suited for situations withs of data , like recordings from previous conversations with customers, or ATM transactions. For example, Google Translate was possible because it “trained” on the vast amount of information on the web, in different languages. In some cases, ML can gain insight or automate decision-making in cases where humans would not be able to. Google search is an example of something that humans can do, but never at the scale and speed at which the Google models are able to show potential answers every time a person types in a query.

AI Subfields

Machine learning is associated with several other artificial intelligence subfields:

Natural language processing

In Natural language processing, machines learn to understand natural language as spoken and written by humans, instead of the data and numbers normally used to program computers. This allows machines to recognize language, understand it, and respond to it, as well as create new text and translate between languages. Examples include chatbots, Siri, Alexa.

Neural Networks

Neural networks are modeled on the human brain, in which thousands/millions of processing nodes are interconnected and organized into layers. In these networks, cells, or nodes are connected, with each cell processing inputs and producing an output that is sent to other neurons. Labeled data moves through the nodes, or cells, with each cell performing a different function.

Deep Learning

Deep learning networks are neural networks with many layers. The layered network can process extensive amounts of data and determine the “weight” of each link in the network. For example, in an image recognition system, some layers of the neural network detect individual features of a face, like eyes, nose, or mouth, while another layer would be able to tell whether those features appear in a way that indicates a face.

Deep learning is modeled on the way the human brain works and powers many machine learning uses, like autonomous vehicles, chatbots, and medical diagnostics. It requires a lot of computing power.

How Machine learning works: Concerns and Challenges

While machine learning can help workers or open new possibilities for businesses, there are several things that one should know about machine learning and its limits.

Explainability

Explainability is the ability to be clear about what the machine learning models are doing and how they make decisions. This is very important because sometimes, systems can be fooled and undermined, or just fail on certain tasks, even those humans can perform easily.

The importance of explaining how a model is working and its accuracy, can vary depending on how it’s being used. For example, a programmer and the viewer would be fine if an algorithm recommending movies is 95% accurate, but that level of accuracy wouldn’t be enough for a self-driving vehicle.

Bias and unintended outcomes

As we all know, machines are trained by humans. Human biases can be incorporated into algorithms and if biased information, or data that reflects existing inequities, is fed to a machine learning program, the program will learn to replicate it and perpetuate forms of discrimination.

Biased outcomes can be avoided by:

• carefully vetting training data and putting organizational support behind ethical artificial intelligence efforts,

• implementing practice of seeking input from people of different backgrounds, experiences, and lifestyles when designing AI systems.

Putting Machine learning to work

Machine learning is changing every industry, and we need to understand the basic principles, the potential, and the limitations before/while implementing it. The way machine learning works for Amazon, or Netflix is not going to translate at a car company or Banks. Each business is different and hence the ML that suits one business type would not work for another.

A basic understanding of ML is as important as finding the right ML to use for the business purpose. This decision should be taken by people with different expertise working together.