|
WO2021069574A1
|
|
Gated linear contextual bandits
|
|
US2021103815A1
|
|
Domain adaptation for robotic control using self-supervised learning
|
|
WO2021058749A1
|
|
Exploration using hypermodels
|
|
US2021089908A1
|
|
Modulating agent behavior to optimize learning progress
|
|
US2021089910A1
|
|
Reinforcement learning using meta-learned intrinsic rewards
|
|
US2021089909A1
|
|
High fidelity speech synthesis with adversarial networks
|
|
WO2021058626A1
|
|
Controlling agents using causally correct environment models
|
|
WO2021058663A1
|
|
Augmenting attention-based neural networks to selectively attend to past inputs
|
|
WO2021058588A1
|
|
Training action selection neural networks using hindsight modelling
|
|
WO2021058578A1
|
|
Fast sparse neural networks
|
|
WO2021058583A1
|
|
Training action selection neural networks using q-learning combined with look ahead search
|
|
US2021078169A1
|
|
Data-driven robot control
|
|
WO2021058270A1
|
|
Gated attention neural networks
|
|
US2021064961A1
|
|
Antisymmetric neural networks
|
|
WO2021009293A1
|
|
Training a neural network to control an agent using task-relevant adversarial imitation learning
|
|
WO2020254400A1
|
|
Robust reinforcement learning for continuous control with model misspecification
|
|
WO2020239641A1
|
|
Hierarchical policies for multitask transfer
|
|
WO2020234449A1
|
|
Generative adversarial networks with temporal and spatial discriminators for efficient video generation
|
|
WO2020234457A1
|
|
Neural network-based memory system with variable recirculation of queries using memory content
|
|
US2020372366A1
|
|
Jointly learning exploratory and non-exploratory action selection policies
|
