论文标题
学习对看不见实体的概括的参数化任务结构
Learning Parameterized Task Structure for Generalization to Unseen Entities
论文作者
论文摘要
现实世界的任务是分层和组成的。任务可以由彼此依赖的多个子任务（或子目标）组成。这些子任务是根据实体（例如“ Apple”，“ Pear”）来定义的，这些实体可以重新组合以形成新的子任务（例如，“ Pickup Apple”和“ Pickup Pear”）。为了有效地解决这些任务，代理必须推断子任务依赖项（例如，代理必须在“将苹果放在锅中”之前执行“拾取苹果”），并将推断的依赖项推广到新的子任务（例如，“将苹果放入锅中的apple与pan Apple”相似）。此外，代理可能还需要解决看不见的任务，这可能涉及看不见的实体。为此，我们制定了参数化子任务图推理（PSGI），这是一种使用子任务实体使用一阶逻辑对子任务依赖性建模的方法。为了促进这一点，我们以零拍的方式学习实体属性，该属性用作量词（例如“ is_pickable（x）”）用于参数化子任务图。我们表明，这种方法比以前的工作更有效地准确地了解了层次和组成任务的潜在结构，并且显示PSGI可以通过在适应过程中看不见的子任务上建模结构来概括。
Real world tasks are hierarchical and compositional. Tasks can be composed of multiple subtasks (or sub-goals) that are dependent on each other. These subtasks are defined in terms of entities (e.g., "apple", "pear") that can be recombined to form new subtasks (e.g., "pickup apple", and "pickup pear"). To solve these tasks efficiently, an agent must infer subtask dependencies (e.g. an agent must execute "pickup apple" before "place apple in pot"), and generalize the inferred dependencies to new subtasks (e.g. "place apple in pot" is similar to "place apple in pan"). Moreover, an agent may also need to solve unseen tasks, which can involve unseen entities. To this end, we formulate parameterized subtask graph inference (PSGI), a method for modeling subtask dependencies using first-order logic with subtask entities. To facilitate this, we learn entity attributes in a zero-shot manner, which are used as quantifiers (e.g. "is_pickable(X)") for the parameterized subtask graph. We show this approach accurately learns the latent structure on hierarchical and compositional tasks more efficiently than prior work, and show PSGI can generalize by modelling structure on subtasks unseen during adaptation.
