Reinforcement Learning

Modified Policy Iteration

Modified Policy Iteration is a reinforcement learning algorithm that blends elements of policy evaluation and policy improvement to efficiently find optimal policies.

Monte Carlo Methods

Monte Carlo Methods are computational algorithms that use random sampling to solve complex mathematical problems and simulate systems with uncertain variables.

Monte Carlo Tree Search

Monte Carlo Tree Search is a heuristic search algorithm used for making decisions in complex environments by combining tree search and random sampling.

Multi-Agent Reinforcement Learning

Multi-Agent Reinforcement Learning is a branch of machine learning where multiple agents learn and interact within a shared environment to achieve individual or collective goals through trial and error.

Nash Equilibrium

Nash Equilibrium is a concept in game theory where no player can benefit by changing their strategy while the other players keep theirs unchanged.

Off-Policy Learning

Off-Policy Learning is a reinforcement learning method where the agent learns the value of an optimal policy independently from the actions taken by a different behavior policy.

On-Policy Learning

On-Policy Learning is a reinforcement learning approach where the agent learns the value of the policy it is currently following to make decisions.

One-Shot Learning

One-shot learning is a machine learning approach where a model learns to recognize or classify something using only a single example.

Option

Option is a contract that gives the buyer the right, but not the obligation, to buy or sell an asset at a predetermined price within a specific time frame.

Partially Observable MDP

Partially Observable MDP is a decision-making framework where an agent makes choices without having full visibility of the current state of the environment.

Perfect Information Game

Perfect Information Game is a type of strategic game where all players have full knowledge of all previous actions and the current state of the game at every decision point.

Policy Iteration

Policy Iteration is a dynamic programming method used in reinforcement learning to find the optimal policy by iteratively evaluating and improving policies.

Prioritized Sweeping

Prioritized Sweeping is a reinforcement learning technique that focuses updates on the most important or impactful states to accelerate learning efficiency.

PUCT

PUCT is a variant of the Upper Confidence Bound algorithm used in Monte Carlo Tree Search to balance exploration and exploitation during decision-making.

Q-Function

Q-Function is a mathematical function used in reinforcement learning to estimate the expected rewards of taking a certain action in a given state.

Q-Learning

Q-Learning is a model-free reinforcement learning algorithm that helps agents learn the best actions to take in specific situations to maximize cumulative rewards.

Q-Value

Q-Value is a statistical measure used to estimate the minimum false discovery rate at which a particular test result can be considered significant.

Real-Time DP

Real-Time DP is a dynamic programming approach that processes data instantly as it becomes available to make immediate decisions or updates.

Regret

Regret is the emotional experience of wishing one had made a different decision or acted differently in the past.

Regret Minimization

Regret minimization is a decision-making strategy that focuses on reducing the negative feeling of regret after making a choice.