Deep Learning (Deep Reinforcement Learning)

Home Glossary Item Deep Learning (Deep Reinforcement Learning)
« Back to Glossary Index

Deep Learning is a subfield of machine learning that focuses on algorithms inspired by the structure and function of the brain called artificial neural networks. It’s capable of processing large volumes of data to identify patterns and make predictions. Deep Learning models are built using multiple layers, “deep” neural networks, where each tier processes input from the preceding layer and passes the output to the subsequent layer. These systems can learn to perform tasks by considering examples, generally without being programmed with rules.


Deep Reinforcement Learning combines Deep Learning and Reinforcement Learning. Reinforcement Learning is a type of machine learning where an agent learns how to behave in an environment, by performing actions and receiving rewards or punishments. Deep Reinforcement Learning applies deep learning techniques to reinforcement learning. It uses neural networks to approximate the reward or value functions, leveraging the layers of the network to represent complicated functions and process high-dimensional inputs.

Deep Reinforcement Learning represents a massive leap in the development of artificial intelligence. It enables machines to make decisions independently and improve their performance based on experiences. The combination of reinforcement learning’s decision-making ability with deep learning’s pattern recognition capabilities results in systems that can adapt and optimize their behavior to achieve specific goals. The most known application of Deep Reinforcement Learning is the development of gaming bots which can outperform human players in complex games.

« Back to Glossary Index