SOTAVerified

Imitation Learning

Imitation Learning is a framework for learning a behavior policy from demonstrations. Usually, demonstrations are presented in the form of state-action trajectories, with each pair indicating the action to take at the state being visited. In order to learn the behavior policy, the demonstrated actions are usually utilized in two ways. The first, known as Behavior Cloning (BC), treats the action as the target label for each state, and then learns a generalized mapping from states to actions in a supervised manner. Another way, known as Inverse Reinforcement Learning (IRL), views the demonstrated actions as a sequence of decisions, and aims at finding a reward/cost function under which the demonstrated decisions are optimal.

Finally, a newer methodology, Inverse Q-Learning aims at directly learning Q-functions from expert data, implicitly representing rewards, under which the optimal policy can be given as a Boltzmann distribution similar to soft Q-learning

Source: Learning to Imitate

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1431–1440 of 2122 papers

TitleStatusHype
Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of Pessimism—0
Introspective Visuomotor Control: Exploiting Uncertainty in Deep Visuomotor Control for Failure Recovery—0
Learning to Simulate on Sparse Trajectory Data—0
Learning from Imperfect Demonstrations from Agents with Varying DynamicsCode0
Optimism is All You Need: Model-Based Imitation Learning From Observation Alone—0
Variational Model-Based Imitation Learning in High-Dimensional Observation Spaces—0
Domain-Robust Visual Imitation Learning with Mutual Information ConstraintsCode1
Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation GapCode1
Decision Making in Monopoly using a Hybrid Deep Reinforcement Learning Approach—0
Generalization Through Hand-Eye Coordination: An Action Space for Learning Spatially-Invariant Visuomotor Control—0
Show:102550
← PrevPage 144 of 213Next →

No leaderboard results yet.