Provably efficient neural GTD algorithm for off-policy learning

Keyphrases

• Policy Learning 100%
• Temporal Difference Algorithm 100%
• Bellman 100%
• Neural Network Architecture 33%
• Primal-dual 33%
• 2-layer 33%
• Approximate Optimal Solution 33%
• Over-parameterized 33%
• Function Approximator 33%
• Neural Network Function 33%
• Min-max Optimization 33%
• ReLU Neural Networks 33%

Mathematics

• Neural Network 100%
• Minimizes 50%
• Approximates 50%
• Min-Max 50%