Online Machine Learning

Online Machine Learning is a method of machine learning in which data becomes available in a sequential order. It is used to update the best predictor for future data at each step, as opposed to batch learning techniques which generate the best predictor by learning on the entire training data set at once.
Online Machine learning is a common technique used in areas of machine learning where it is computationally infeasible to train over the entire dataset, requiring the need of out-of-core algorithms. It is also used in situations where it is necessary for the algorithm to dynamically adapt to new patterns in the data, or when the data itself is generated as a function of time, e.g., stock price prediction.

What is Online Machine Learning ?

It is called online machine learning because the ML programming must be connected to a network that provides a dynamic input stream.
Online learning algorithms are often compared to batch learning algorithms. Batch learning algorithms are static and use large datasets to train and test outcomes before the model makes production. In contrast, online learning algorithms are trained incrementally as new data becomes available.
Basic ML programs took a fixed set of input data from a file e.g, CSV, and then worked on it sequentially. With online machine learning, it is different data channel . Online machine learning program may be taking in real-time information from sensors in a manufacturing environment, or text input from users over the internet, or something else that comes in as input in real time. The machine learning program is made to deal with these real-time data streams and produce results. Online machine learning can help make machine learning algorithms more capable in many fields.

Training and Computational Complexity

• Training in online ML, update in hyperparameters happen at every step especially depending on the current sample under observation and also the current state of the model. The model is continuously seeing new data and improving itself.
• Where data comes in the stream like big data, our online model is ready to update and tune itself. This may sometimes result in more cost to training because there will be a requirement of a lot of resources to train the model constantly.
• Online models are difficult to maintain after deployment because the model has to converge all the data points in real-time any changes and ambiguities in data will affect the overall performance and prediction of the model. Therefore it is necessary to monitor the process throughout the time.

Advantages of Online Machine Learning

One major benefit of online learning is adaptability. The model is able to adjust and learn from datasets with different patterns and distributions as they come. And also, online learning does not require so much memory for storing data. Once the model has been trained over a specific observation, there is no need to store it.

Disadvantages of Online Machine Learning

A drawback of online learning is the complexity behind implementing. Because learning takes place on the net, we have to consider how the model will be updated and how the data will be processed just to name a few. Ultimately this requires more resources hence capital intensive.

Online ML vs Offline ML

Recommended for you:
Batch Learning vs Online Learning
Genetic Algorithm in Machine Learning