Author name: tirta

Deep Q-Network (DQN) is a valuable reinforcement learning algorithm that utilizes a neural network to estimate the action-value function. The […]

A heuristic in congestion control is a rule-based approach used to guide data transmission rates and manage network congestion. It

Representativeness heuristic is a cognitive shortcut where individuals assess the likelihood of an event based on its perceived similarity to

Reinforcement learning (RL) involves defining the state and action spaces for an agent interacting with an environment. State parametrization defines

One-step reinforcement learning bandit is a simple yet effective algorithm for solving sequential decision problems. In this setting, the agent

Iterative learning control (ILC) improves the performance of closed-loop control systems by utilizing feedback to enhance their accuracy over time.

RL Parametrizations: Reinforcement learning (RL) involves designing parametrized policies and value functions to represent the agent’s behavior and estimate the

Stylized Offline Reinforcement Learning combines the principles of reinforcement learning with offline data utilization. It leverages pre-collected data to train

Optimal control dynamic programming is a mathematical approach to finding the optimal control policy for a dynamic system. It involves

Variable state space in reinforcement learning refers to environments where the state space—the set of possible states that an agent

Economic Model Predictive Control (EMPC) is a sophisticated technique that combines economic principles and optimization with Model Predictive Control (MPC).

Reward Booster Reinforcement Learning is a reinforcement learning method that uses a reward booster to encourage the agent to explore