PAC-Bayes Control: Synthesizing Controllers that Provably Generalize to Novel Environments

Computer Science

• Convex Optimization 100%
• Depth Measurement 100%
• Finite Control 100%
• Learning System 100%
• Machine Learning 100%
• Obstacle Avoidance 100%
• Optimization Problem 100%
• Relative Entropy 100%
• Robot 100%
• Robotics 100%
• Sensor Measurement 100%
• Stochastic Gradient Descent 100%
• Supervised Learning 100%
• Synthesis Algorithm 100%

Keyphrases

• Bayes 100%
• Continuous Action Space 25%
• Continuous State Space 25%
• Control Policy 25%
• Controller 100%
• Controller Parameterization 25%
• Convex Optimization 25%
• Depth Measurement 25%
• Expected Cost 25%
• Finite Control 25%
• Generalization Guarantees 25%
• Generalized Theory 25%
• High Probability 25%
• Machine Learning 25%
• Nonlinear Dynamics 25%
• Novel Environment 100%
• Obstacle Avoidance 25%
• Optimization Problem 25%
• Partially Observable 25%
• Policy Space 25%
• Probably Approximately Correct 100%
• Relative Entropy Programming 25%
• Robotic System 25%
• Sensor Measurement 25%
• Stochastic Controller 25%
• Stochastic Gradient Descent 25%
• Supervised Learning 25%
• Synthesis Algorithm 25%
• Synthetic Control Method 25%