Mathematics
Regret
100%
Learning
58%
Unknown
33%
Efficient Algorithms
31%
Reinforcement Learning
29%
On-line Control
28%
Linear Quadratic Control
27%
Learning Algorithm
25%
Policy
22%
Costs
22%
Linear Quadratic Regulator
21%
Mixing Time
19%
Markov Decision Process
18%
Linear Dynamical Systems
18%
Optimal Rates
18%
Linear Control
18%
Function Approximation
18%
Online Algorithms
16%
Shortest path
16%
Cost Function
15%
Paradigm
15%
Controller
14%
Computational Complexity
14%
Square root
14%
Uncertainty
13%
Logarithmic
12%
Trajectory
12%
Online Learning
11%
Transition Matrix
11%
Horizon
11%
Optimal Policy
10%
Quadratic Loss
10%
Steady-state Distribution
9%
Oracle
9%
Reward
9%
Upper bound
9%
Learning Rate
7%
Regression
7%
Linear Time
6%
State Transition
6%
Quadratic Systems
6%
Linear Optimization
6%
Dynamic Environment
6%
Finite Horizon
6%
Mirror
5%
Nondegeneracy
5%
Wages
5%
Schedule
5%
Loss Function
5%
Descent
5%
Engineering & Materials Science
Reinforcement learning
45%
Learning algorithms
37%
Feedback
27%
Markov processes
27%
Costs
23%
Convex optimization
22%
Control systems
19%
Relaxation
14%
Planning
12%
Cost functions
11%
Sampling
10%
Dynamical systems
10%
Markov chains
10%
Trajectories
8%
Polynomials
7%
Wages
6%
Controllers
6%
Computational complexity
6%
Random processes
5%
Random variables
5%
