Deep reinforcement learning
Deep reinforcement learning is reinforcement learning that uses deep neural networks to represent policies, value functions, or other decision-making components.
Applied Reinforcement Learning
A reinforcement-learning approach that uses deep neural networks to approximate policies, value functions, or related decision components.
- How it is used
- Frames high-dimensional sequential decisions as problems that neural-network function approximation can address.
- How deep it goes
- Explains neural networks as value functions, policies, or both, and frames algorithms as components wired into a loop.
- Examples
- Uses uncertain tumor-dose adjustments and an eight-choice allocation policy for a six-ETF portfolio.
- What you can do after
- Understand algorithm differences by identifying which shared component each method changes.
Read how Applied Reinforcement Learning explains deep reinforcement learning →
Evolutionary Deep Learning: Genetic algorithms and neural networks
Deep reinforcement learning is reinforcement learning that uses deep neural networks to represent policies, value functions, or related decision-making components.
- How it is used
- Solves reinforcement-learning problems whose state and action spaces make tables impractical, including classic control and game environments.
- How deep it goes
- Explains neural-network substitution and Q-learning supervision, then stops short of a detailed network-training implementation.
- Examples
- Uses CartPole, Acrobot, MountainCar, LunarLander, Atari games, action spaces, observation spaces, rewards, and rendered episodes.
- What you can do after
- Apply via OpenAI Gym by configuring environments, inspecting spaces, running episodes, collecting rewards, and rendering results.
Also covered in
What to know first
- actionAn action is a choice or operation selected or performed by an actor that can affect a system, environment, or process.
- environmentAn environment is the external context, system, or conditions within which an agent or application operates and interacts.
- stateState is information describing the condition or configuration of a system or computation at a particular time that may influence its behavior or subsequent transitions.
- reinforcement learningReinforcement learning is a machine-learning paradigm in which an agent learns behavior or a decision policy through interaction with an environment and evaluative feedback, such as rewards.
Related concepts
- Q-learningcontrasts with · Q-learning is a model-free reinforcement-learning algorithm that estimates state-action values through updates based on rewards and subsequent state values.
- language modeluses · A language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- reinforcement learningcontrasts with · Reinforcement learning is a machine-learning paradigm in which an agent learns behavior or a decision policy through interaction with an environment and evaluative feedback, such as rewards.
- causal structureuses · A causal structure is an organization of variables and directed causal relationships that describes how factors influence outcomes.
- DQNincludes · A deep Q-network (DQN) is a reinforcement-learning method that uses a neural network to approximate the state-action value function in Q-learning.
- neuroevolution of augmenting topologiesused by · NeuroEvolution of Augmenting Topologies (NEAT) is an evolutionary method that evolves neural-network weights and connection topologies, typically through genomes and speciation.
Questions about deep reinforcement learning
What is deep reinforcement learning?
Deep reinforcement learning is reinforcement learning that uses deep neural networks to represent policies, value functions, or other decision-making components.
Which Manning books explain deep reinforcement learning?
Applied Reinforcement Learning and Evolutionary Deep Learning: Genetic algorithms and neural networks.
What should I know before learning deep reinforcement learning?
Manning books that teach deep reinforcement learning usually introduce action, environment, state and reinforcement learning first.
What is the difference between deep reinforcement learning and Q-learning?
Deep reinforcement learning is reinforcement learning that uses deep neural networks to represent policies, value functions, or other decision-making components. Q-learning is a model-free reinforcement-learning algorithm that estimates state-action values through updates based on rewards and subsequent state values.
What kinds of deep reinforcement learning are there?
Kinds of deep reinforcement learning include DQN.
What is deep reinforcement learning also called?
Deep reinforcement learning is also called Deep RL.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

