The Robot Practiced Where It Had Failed

Uniform demonstration collection wastes operator time once failures concentrate in a small part of the task. Mulligan makes the initial-state distribution itself a decision, sending new episodes toward observed failures and untried placements. It also trains a value function on all data, including failed runs that imitation learning normally discards. Across three real robot tasks evaluated on 2,550 blinded held-out episodes and two simulated tasks, the combined system improved final real-task success by 10 to 34 percentage points over uniform collection at matched budgets. Human intervention kept 98% of collection episodes productive. The method treats failure locations as a scarce-data map rather than an embarrassment to average away.