What is Competitive Learning?
Competitive learning is a type of unsupervised learning model used in machine learning and artificial intelligence systems. Some of the exciting new formats of machine learning projects are based in part on competitive learning, including self-organizing components of neural networks. That makes this an integral idea in the machine learning community.
In a competitive learning model there are hierarchical sets of units in the network with inhibitory and excitatory connections. The excitatory connections are between individual layers and the inhibitory connections are between units in layered clusters. Units in a cluster are either active or inactive.
Scientists explain that the configuration of active units is an input pattern that is sent to the next level. In processes like vector quantization, professionals can see the principles of competitive learning at work. Competitive learning also exists alongside other learning models such as ensemble learning, where more than one learning unit works together to produce one result in a separate effort.