In
machine learning
Machine learning (ML) is a field of inquiry devoted to understanding and building methods that 'learn', that is, methods that leverage data to improve performance on some set of tasks. It is seen as a part of artificial intelligence.
Machine ...
, systems which employ offline learning do not change their approximation of the target function when the initial training phase has been completed. These systems are also typically examples of
eager learning In artificial intelligence, eager learning is a learning method in which the system tries to construct a general, input-independent target function during training of the system, as opposed to lazy learning, where generalization beyond the training ...
.
While in online learning, only the set of possible elements is known, in offline learning, the identity of the elements as well as the order in which they are presented is known to the learner.
Applications for robotics control
The ability of robots to learn is equal to create a
table (information)
A table is an arrangement of information or data, typically in rows and columns, or possibly in a more complex structure. Tables are widely used in communication, research, and data analysis. Tables appear in print media, handwritten notes, comp ...
which is filled with values. One option for doing so is
programming by demonstration. Here, the table is filled with values by a human teacher. The demonstration is provided either as
direct numerical control
Direct numerical control (DNC), also known as distributed numerical control (also DNC), is a common manufacturing term for networking CNC machine tools. On some CNC machine controllers, the available memory is too small to contain the machining p ...
policy which is equal to a trajectory, or as an indirect
objective function
In mathematical optimization and decision theory, a loss function or cost function (sometimes also called an error function) is a function that maps an event or values of one or more variables onto a real number intuitively representing some "cost ...
which is given in advance.
Offline learning is working in
batch mode. In step 1 the task is demonstrated and stored in the table, and in step 2 the task is reproduced by the robot. The pipeline is slow and inefficient because a delay is there between behavior demonstration and skill replay.
A short example will help to understand the idea. Suppose the robot should learn a
wall following task and the internal table of the robot is empty. Before the robot gets activated in the replay mode, the human demonstrator has to teach the behavior. He is controlling the robot with
teleoperation
Teleoperation (or remote operation) indicates operation of a system or machine at a distance. It is similar in meaning to the phrase "remote control" but is usually encountered in research, academia and technology. It is most commonly associat ...
and during the learning step the skill table is generated. The process is called offline, because the robot control software is doing nothing but the device is utilized by the human operator as a
pointing device
A pointing device is a human interface device that allows a user to input spatial (i.e., continuous and multi-dimensional) data to a computer. CAD systems and graphical user interfaces (GUI) allow the user to control and provide data to ...
for driving along the wall.
See also
*
Online learning, the opposite model
*
Incremental learning
In computer science, incremental learning is a method of machine learning in which input data is continuously used to extend the existing model's knowledge i.e. to further train the model. It represents a dynamic technique of supervised learnin ...
, a learning model for the incremental extension of knowledge
References
Machine learning
{{compu-stub