time horizon
Time horizon refers to the timeframe over which predictions or decisions are made in AI applications, particularly in reinforcement learning. It helps define the scope of learning, planning, and expected outcomes for agents interacting with their environments.
- A Bayesian Approach to Contextual Dynamic Pricing using the Proportional Hazards Model with Discrete Price Data
- Bandit and Delayed Feedback in Online Structured Prediction
- Finite Sample Analyses for Continuous-time Linear Systems: System Identification and Online Control
- Instance-Dependent Regret Bounds for Nonstochastic Linear Partial Monitoring
- Measuring AI Ability to Complete Long Software Tasks
- When Lower-Order Terms Dominate: Adaptive Expert Algorithms for Heavy-Tailed Losses