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 1701–1725 of 2122 papers

TitleStatusHype
Plan Your Target and Learn Your Skills: State-Only Imitation Learning via Decoupled Policy Optimization—0
Playful Interactions for Representation Learning—0
Play it by Ear: Learning Skills amidst Occlusion through Audio-Visual Imitation Learning—0
PLOP: Probabilistic poLynomial Objects trajectory Planning for autonomous driving—0
Policy Agnostic RL: Offline RL and Online RL Fine-Tuning of Any Class and Backbone—0
Policy Architectures for Compositional Generalization in Control—0
Policy Contrastive Imitation Learning—0
Policy Decorator: Model-Agnostic Online Refinement for Large Policy Model—0
Policy Gradient Bayesian Robust Optimization for Imitation Learning—0
Policy Improvement via Imitation of Multiple Oracles—0
Policy Optimization by Genetic Distillation—0
Policy Optimization by Local Improvement through Search—0
Policy Regularization on Globally Accessible States in Cross-Dynamics Reinforcement Learning—0
Policy Search for Motor Primitives in Robotics—0
Population-Guided Imitation Learning—0
Positive-Unlabeled Reward Learning—0
Practical Imitation Learning in the Real World via Task Consistency Loss—0
Precise Affordance Annotation for Egocentric Action Video Datasets—0
Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning—0
Prediction with Action: Visual Policy Learning via Joint Denoising Process—0
Predictive Maneuver Planning with Deep Reinforcement Learning (PMP-DRL) for comfortable and safe autonomous driving—0
Predictive Modeling of Periodic Behavior for Human-Robot Symbiotic Walking—0
Predictive Red Teaming: Breaking Policies Without Breaking Robots—0
Predictive-State Decoders: Encoding the Future into Recurrent Networks—0
Preference-conditioned Pixel-based AI Agent For Game Testing—0
Show:102550
← PrevPage 69 of 85Next →

No leaderboard results yet.