Lecture

Reinforcement Learning: Q-Learning

Graph Chat

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

Description

This lecture introduces the concept of reinforcement learning, focusing on Q-Learning as a model-free method for agents to learn by trial-and-error. It covers the exploration-exploitation trade-off, action values, and policies. The lecture also discusses the convergence properties of Q-Learning, the importance of human-compatible AI, and the impact of algorithms on society.

Official source

Related concepts (32)

Reinforcement learning

Reinforcement learning (RL) is an area of machine learning concerned with how intelligent agents ought to take actions in an environment in order to maximize the notion of cumulative reward. Reinforcement learning is one of three basic machine learning paradigms, alongside supervised learning and unsupervised learning. Reinforcement learning differs from supervised learning in not needing labelled input/output pairs to be presented, and in not needing sub-optimal actions to be explicitly corrected.

Q-learning

Q-learning is a model-free reinforcement learning algorithm to learn the value of an action in a particular state. It does not require a model of the environment (hence "model-free"), and it can handle problems with stochastic transitions and rewards without requiring adaptations. For any finite Markov decision process (FMDP), Q-learning finds an optimal policy in the sense of maximizing the expected value of the total reward over any and all successive steps, starting from the current state.

Split-brain

Split-brain or callosal syndrome is a type of disconnection syndrome when the corpus callosum connecting the two hemispheres of the brain is severed to some degree. It is an association of symptoms produced by disruption of, or interference with, the connection between the hemispheres of the brain. The surgical operation to produce this condition (corpus callosotomy) involves transection of the corpus callosum, and is usually a last resort to treat refractory epilepsy.

Brain damage

Neurotrauma, brain damage or brain injury (BI) is the destruction or degeneration of brain cells. Brain injuries occur due to a wide range of internal and external factors. In general, brain damage refers to significant, undiscriminating trauma-induced damage. A common category with the greatest number of injuries is traumatic brain injury (TBI) following physical trauma or head injury from an outside source, and the term acquired brain injury (ABI) is used in appropriate circles to differentiate brain injuries occurring after birth from injury, from a genetic disorder (GBI), or from a congenital disorder (CBI).

Deep reinforcement learning

Deep reinforcement learning (deep RL) is a subfield of machine learning that combines reinforcement learning (RL) and deep learning. RL considers the problem of a computational agent learning to make decisions by trial and error. Deep RL incorporates deep learning into the solution, allowing agents to make decisions from unstructured input data without manual engineering of the state space. Deep RL algorithms are able to take in very large inputs (e.g.

Related lectures (64)

Landscape and Generalisation in Deep Learning

Explores the challenges and insights of deep learning, focusing on loss landscape, generalization, and feature learning.

Reinforcement Learning: TD Learning and SARSA Variants

Discusses reinforcement learning, focusing on temporal difference learning and SARSA algorithm variations.

AI for Science

Explores fundamental principles in scientific research, the impact of computers, numerical algorithms, and deep learning in solving high-dimensional problems.

Data-Driven Modeling in Neuroscience: Meenakshi Khosla

By Meenakshi Khosla explores data-driven modeling in large-scale naturalistic neuroscience, focusing on brain activity representation and computational models.

Deep Q-Learning: DeepRL1.1