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 1476–1500 of 2122 papers

TitleStatusHype
Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations—0
Learning to Infer User Interface Attributes from Images—0
Learning to Interactively Learn and Assist—0
Learning to Make Decisions via Submodular Regularization—0
Learning to Multi-Task Learn for Better Neural Machine Translation—0
Learning to Optimize in Model Predictive Control—0
Learning to Optimize Permutation Flow Shop Scheduling via Graph-based Imitation Learning—0
Learning to Prove Trigonometric Identities—0
Learning to Reach Goals Without Reinforcement Learning—0
Learning to Reason in Large Theories without Imitation—0
Learning to request guidance in emergent language—0
Learning to Request Guidance in Emergent Communication—0
Learning to Search for Fast Maximum Common Subgraph Detection—0
Learning to Search in Branch and Bound Algorithms—0
Learning to Search via Retrospective Imitation—0
Learning to Select Nodes in Bounded Suboptimal Conflict-Based Search for Multi-Agent Path Finding—0
Learning to Simulate on Sparse Trajectory Data—0
Learning to Structure an Image with Few Colors and Beyond—0
Learning to Superoptimize Real-world Programs—0
Learning to Transfer In-Hand Manipulations Using a Greedy Shape Curriculum—0
Learning Translations: Emergent Communication Pretraining for Cooperative Language Acquisition—0
Learning Value Functions from Undirected State-only Experience—0
Learning Vision-based Flight in Drone Swarms by Imitation—0
Learning What's Easy: Fully Differentiable Neural Easy-First Taggers—0
Learning with Imperfect Models: When Multi-step Prediction Mitigates Compounding Error—0
Show:102550
← PrevPage 60 of 85Next →

No leaderboard results yet.