Patent Leaderboard
USPTO Patent Rankings Data through Dec 31, 2025
Home › Meta-gradient updates for training return functions for reinforcement learning systems

Meta-gradient updates for training return functions for reinforcement learning systems

US Patent 11836620 · Granted Dec 5, 2023

Estimated economic value: $99,298,000 ⓘ

Assignee

Inventors

View full patent text on Google Patents →

More Patents by These Inventors

Continual reinforcement learning with a multi-task agent2024
Training action selection neural networks using look-ahead search2024
Environment prediction using reinforcement learning2024
Environment prediction using reinforcement learning2024
Training neural networks using a prioritized experience memory2024
Multi-agent reinforcement learning with matchmaking policies2024
Reinforcement learning using baseline and policy neural networks2024
Training reinforcement learning neural networks2024
Reinforcement learning with auxiliary tasks2023
Surfboard fin2023
Training action selection neural networks using look-ahead search2023
Continuous control with deep reinforcement learning2023