sequential task

multi-task learning (2575) reinforcement learning (4352) throughput increase (1) low regret (3) action representation (16) generalization objective (1) latency reduction (77) zero-shot generalization (194) regret bound (1926) reward mapping (1)