NTHRYS › PhD Assistance › Reinforcement Learning

Reinforcement Learning

Field

▾

Reinforcement Learning

Select a category to explore research frontiers

Loading categories...

Research Frontiers in Safe Reinforcement Learning with Constraints

Development of RL methods that guarantee safety constraints and prevent dangerous behaviors during both training and deployment in critical applications.

Constraint Emergence in Hierarchical Multi-Agent Learning

Specification Gaming and the Reward Robustness Frontier

Safety-Critical Sim-to-Real Transfer Under Distribution Shift

Formal Verification of Neural Policy Guarantees

Intrinsic Motivation Within Hard Constraint Boundaries

Adversarial Robustness in Constrained Action Spaces

Human Preference Elicitation for Implicit Constraint Discovery

Causal Reasoning Under Safety-Critical Constraints

Constraint Relaxation and Policy Adaptation at Deployment

Uncertainty Quantification in Safe Offline Reinforcement Learning

All Reinforcement Learning PhD categories