reward function
A reward function is a mathematical description of the incentives given to an agent in reinforcement learning. It quantifies the objective of the learning task, guiding the agent’s actions towards desirable outcomes by providing feedback on their performance.
- AutoEdit: Automatic Hyperparameter Tuning for Image Editing
- Avoiding exp(R) scaling in RLHF through Preference-based Exploration
- Beyond Greedy Exits: Improved Early Exit Decisions for Risk Control and Reliability
- Capturing Individual Human Preferences with Reward Features
- Doubly Robust Alignment for Large Language Models
- Iterative Foundation Model Fine-Tuning on Multiple Rewards
- Near-Optimal Experiment Design in Linear non-Gaussian Cyclic Models
- Non-Stationary Lipschitz Bandits
- Reasoning as an Adaptive Defense for Safety
- Reducing the Probability of Undesirable Outputs in Language Models Using Probabilistic Inference
- SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learning
- Self-Improving Embodied Foundation Models
- Trust Region Reward Optimization and Proximal Inverse Reward Optimization Algorithm
- VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning