Domain Randomization

Domain Randomization simply uniformly samples tasks from the task space. It can be a strong baseline in environments with relatively small task spaces.

class syllabus.curricula.domain_randomization.BatchedDomainRandomization(batch_size: int, task_space, warmup_batches: int = 5, **kwargs)

Bases: Curriculum

A simple but strong baseline for curriculum learning that uniformly samples a task from the task space.

sample(k: int = 1) → Any

Sample k tasks from the curriculum.

Parameters:

k – Number of tasks to sample, defaults to 1

Returns:

Either returns a single task if k=1, or a list of k tasks

update_on_episode(episode_return, length, task, progress, env_id: int | None = None) → None

Update the curriculum with episode results from the environment.

Parameters:

• episode_return – Episodic return

• length – Length of the episode

• task – Task for which the episode was completed

• progress – Progress toward completion or success rate of the given task. 1.0 or True typically indicates a complete task.

• env_id – Environment identifier

class syllabus.curricula.domain_randomization.DomainRandomization(task_space: TaskSpace, random_start_tasks: int = 0, task_names: Callable | None = None, record_stats: bool = False)

Bases: Curriculum

A simple but strong baseline for curriculum learning that uniformly samples a task from the task space.

class syllabus.curricula.domain_randomization.SyncedBatchedDomainRandomization(batch_size: int, task_space, warmup_batches: int = 1, uniform_chance: float = 0.05, **kwargs)

Bases: Curriculum

A simple but strong baseline for curriculum learning that uniformly samples a task from the task space.

sample(k: int = 1) → Any

Sample k tasks from the curriculum.

update_batch()

Update the current batch.